We’ve written time and again about how QR codes are used in phishing schemes. Our secure email gateway solution even includes technology to read these codes — not just from emails, but also from attachments — and check the embedded links. Yet, attackers haven’t given up on trying to send QR codes to their victims. Lately, we’ve increasingly seen them use ASCII art for this purpose — images composed of characters. This seems particularly ironic considering that phishers once tried to evade link scanning by hiding links in images, and now they’re trying to dodge image scanning by going back to text. But with a few twists.
The lost art of ASCII, and how attackers use it
It’s hard to believe today, but there was a time when computers couldn’t display graphics. Consequently, the very first computer images were constructed from text characters. Following the adoption of the standard in 1963, characters from the ASCII (American Standard Code for Information Interchange) set were used for this type of artwork to ensure that images looked the same across different computers. Over time, other text symbols (for example, from the extended Unicode set) began to be used to create images, but the name “ASCII graphics” remained the term used to describe this art form as a whole. There were serious artists working in this medium, the earliest websites were designed with ASCII art, and even the first computer pornography was rendered with text characters.
As image display technology evolved, ASCII art began to fall out of fashion. It saw a major resurgence in the 2000s during the heyday of email spam. Back then, spammers primarily used it because it allowed them to disguise blatant spam keywords that could trigger mail filters, while also placing less load on mail servers than images. Additionally, since many users paid for volume of internet traffic at the time, they often disabled image loading in their email clients. Naturally, at that time, we augmented our email security solutions with technology specifically designed to block ASCII art.
Now, ASCII art has been rediscovered — this time by those looking to bypass technology that recognizes QR codes within images.
What does ASCII art phishing look like?
Here’s a recent example. The pretext itself is pretty run-of-the-mill: someone has supposedly sent to victim a confidential document via DocuSign, but to open it the recipient needs to scan the QR code in the email to visit a website and enter corporate login credentials.
A QR code rendered with unicode characters. We’ve blurred out a portion of the code to prevent the malicious link from being scanned.
Admittedly, the code looks weird. This is primarily because it’s drawn piece-by-piece in pseudo-graphic elements, and even the gaps between the lines can be seen. In reality, there’s no actual image in the e-mail message code; the QR code looks something like this behind the scenes:
ASCII art inside the email code
As a result, link scanners can’t see the link, and image analysis tools can’t find the URL hidden inside the QR code, so the attackers assume the phishing email is going to reach the victim just fine. Spoiler alert: no, we haven’t forgotten how to block ASCII art.
Is a QR code in an email even normal?
In theory, there are situations where using a QR code makes sense. It’s a fairly convenient way to share contacts, a link to a mobile app, a map location, or a configuration. In other words, it works well whenever information needs to be delivered specifically to the recipient’s mobile device.
However, someone using a QR code to make you enter corporate credentials on a mobile device is an instant red flag. And when that QR code is generated with ASCII art, it’s clearly a phishing attempt or an effort to lure you to a malicious URL. This trick can only have one purpose — an attempt to bypass security controls.
Additionally, we recommend regular security awareness training to educate employees on modern phishing tactics. Specifically, to explain that ASCII art in modern emails can be a telltale sign of an attempted phishing attack.
Netflix, Apple TV+, Disney+, Hulu, Amazon Prime, YouTube Premium… The average law-abiding family today pays for five to 10 subscriptions just to watch their shows of choice, with the monthly bill easily crossing the hundred-dollar mark. It’s no surprise, then, that social media and online marketplaces are seeing a surge in demand for the “magic boxes” that popped up at the end of 2025: Android-powered TV boxes that promise to unlock thousands of channels and every streaming service subscription-free for a one-time purchase.
Ads for these devices are flooding TikTok and Instagram: smiling influencers unbox the SuperBoxes, plug them into a TV, and browse endlessly through channels. It looks like the ultimate life hack against subscription fatigue, right? In reality, it’s one of the easiest ways to invite a botnet into your home network.
A promotional video on TikTok explaining how great it is when the cheese is free you can just go ahead and cancel all your subscriptions
What’s wrong with these cheap TV boxes?
Stories about malicious TV boxes have surfaced before, but right now, their marketing has reached a truly alarming scale.
At the end of 2025, analysts examined several models of the popular SuperBox device available from major retail stores and online marketplaces. The findings were deeply concerning: immediately upon powering up, the devices began pinging the servers of the Chinese messaging app Tencent QQ, as well as the Grass proxy service — effectively renting out the owner’s internet bandwidth to third parties.
Inside the firmware, researchers discovered applications completely uncharacteristic of a media player: a network scanner, a traffic analyzer, and tools for DNS hijacking. Consequently, the device not only streams pirated content but also scans the local network for other targets (including industrial SCADA interfaces), and stands ready to participate in DDoS attacks. The SuperBoxes were also found to contain folders with the telltale name “secondstage”, a textbook indication of multi-stage malware.
More recently, in April 2026, the Darknet Diaries podcast featured an interview with a security researcher known by the alias D3ada55, who shared plenty of intriguing details about these boxes — including the fact that they were still openly sold on major platforms like Amazon, Walmart, and Best Buy.
The infection chronicles: BADBOX to Keenadu
The SuperBox case is far from the only instance where Android devices have been turned into botnet nodes — or sold infected right out of the box. Here’s a look at the most recent cases:
BADBOX 2.0. In July 2025, Google filed a lawsuit against the operators of a botnet that compromised over 10 million Android devices — mostly cheap TV boxes, tablets, and projectors lacking Google Play Protect certification. As we reported earlier, BADBOX 2.0 specifically targets TV boxes, operating simultaneously as a proxy network and an ad fraud engine.
Kimwolf. In December 2025, the QiAnXin XLab team uncovered a DDoS botnet that had hijacked around 1.8 million Android devices. The infected hardware included generic models from off-brand manufacturers sporting high-profile names like TV BOX, SuperBox, XBOX, SmartTV, and others. The infection footprint was massive, with compromised devices shipped worldwide. Among the hardest-hit countries were Brazil, India, the U.S., Argentina, South Africa, the Philippines, and Mexico.
Keenadu. Our experts discovered this malware lurking in the firmware of brand-new devices back November 2025, though it didn’t gain widespread attention until after we published a study about it in February 2026. Keenadu masquerades as legitimate system components, embedding itself even into facial-recognition unlock apps, potentially granting attackers access to biometrics, banking data, and personal messages.
All of these stories share the same origin: the Triada Trojan, first documented by our researchers back in 2016 and dubbed at the time “one of the most advanced mobile Trojans”. Over the past decade it has evolved from a standard piece of malware into a modular backdoor baked directly into firmware during manufacturing.
How the infection scheme works
Manufacturers of cheap TV boxes cut corners on absolutely everything: Google Play Protect certification, firmware audits, and security updates. Many of these devices run on the Android Open Source Project without any security guarantees whatsoever. Somewhere along the supply chain — whether at the factory, through a middleman, or at a distributor — a backdoor gets injected into the firmware image. Our experts suspect that the manufacturer itself might not even be aware of the compromise.
The sheer scale of the infection turns millions of identical boxes into the perfect foundation for a botnet: every compromised device represents a unique IP address that can be rented out to anyone. Botnet operators like Kimwolf monetize this not only through distributed DDoS attacks but also by reselling the bandwidth of infected smart TVs and streaming boxes.
What this means for you
An infected TV box sits right in your living room, connected to your home Wi-Fi. That means it can see smartphones running banking apps, network-attached storage (NAS) units holding family archives, IP cameras, smart locks, work laptops, and any other the devices connected to your Wi-Fi network.
With this kind of beachhead inside your home network, an attacker can intercept unencrypted traffic, spoof DNS requests, scan ports, and hunt for vulnerabilities on neighboring devices. On top of that, they can use your IP address for fraudulent activity. As a result, in the best-case scenario, your IP will end up blacklisted, and legitimate services will start blocking you for suspicious activity; in the worst-case scenario, law enforcement could come knocking on your door.
How to spot a potentially dangerous gadget
You should be on alert if a device:
Is sold under a no-name brand like T95, X96Q, MX10, TV BOX, SuperBox, or some such
Promises free lifetime access to paid premium services for a one-time fee
Requires you to disable Google Play Protect, or install third-party APK files during the initial setup
Lacks Play Protect certification entirely
Is promoted through aggressive spam campaigns on social media
How to avoid hosting a botnet node
Buy certified TV boxes that feature Google Play Protect, or purchase devices directly from reputable telecom operators and internet service providers.
Isolate all smart home devices. Set up a separate Wi-Fi network on your home router for TV boxes, cameras, smart speakers, robot vacuums, and similar gear, while keeping smartphones, NAS units, and computers on the main network. This prevents malware from spreading to your critical gadgets.
Regularly update the firmware on all your devices, and don’t forget about your router — it’s another vulnerable link in the chain.
Remove any applications from your Android TV box that you didn’t install yourself, especially alternative app stores, Wi-Fi “boosters”, and “system cleaners”.
Monitor your traffic. Modern routers and Kaspersky Premium can display which devices are connecting to where. Frequent connections from a media player to servers in China are a major security red flag.
Install Kaspersky Premiumon all your devices — it protects against Trojans, and blocks the phishing pages often used to distribute infected APK files.
Don’t disable Google Play Protect, and avoid installing APKs from shady sources — this is the primary infection vector that bypasses the official app store.
If in doubt, return the TV box. A cheap streaming device isn’t worth risking your biometrics, banking data, or the reputation of your IP address.
Want to know how else to protect your smart home devices? Read more in our related posts:
While many companies are intentionally rolling out AI to boost quality and efficiency, unsanctioned AI tools are cropping up in corporate environments even faster. Software vendors are baking AI right into products companies already use (think Microsoft Copilot and Google Gemini), while employees are taking matters into their own hands and installing tools on the sly. As a result, businesses are staring down a poorly managed data leak channel: staff paste information from corporate systems into AI chatbots, sending data not just to the SaaS vendor, but straight to the developers behind the underlying AI model. Both the risks and the mitigation strategies vary depending on the type of AI system in play. We break down this broad topic, focusing heavily on tools for spotting and blocking AI at two distinct levels.
Types of unwanted AI systems
Depending on the type of AI in question, managing and blocking its use requires a different playbook. It’s essential to break down AI into four distinct categories:
Platform-native AI capabilities. Think Microsoft Copilot, Google Gemini, and Apple Intelligence, along with AI features baked right into browsers. The tricky thing about these is that they’re built into everyday essentials, are instantly available to every user (sometimes popping up aggressively), and most importantly, vendors try to turn them on by default.
AI companions embedded in business apps. This bucket includes Slack AI, Zoom AI Companion, Notion AI, Jira’s Rovo assistant, and the like. These are tied to a single application and are completely inseparable from it.
Standalone web and app-based chatbots. ChatGPT, Claude, Perplexity, Character AI, local setups like LM Studio, browser extensions, and agentic browsers like Comet. Apps and services in this category are usually adopted by employees on their own without permission: classic examples of shadow AI.
Desktop-native multi-functional agents. This group features tools like OpenClaw, NanoClaw, NemoClaw, and others. They pose the biggest threat because they come with broad access rights by default and actively process untrusted data from the open web.
How to deal with unwanted AI
Every company, depending on its industry, appetite for innovation, and risk tolerance, needs to draw its own line in the sand between recommended, approved case-by-case, and completely banned use cases for specific AI products. Regulated sectors like healthcare play by one set of rules, while retail businesses operate under an entirely different playbook. Either way, after analyzing exactly which AI tools have already slipped into the organization, corporate policies need to be fine-tuned. That’s why the first order of business is employing existing infosec and logging tools to scan corporate infrastructure.
Depending on the chosen strategy, the uncovered AI systems can be:
Disabled or restricted by using the built-in corporate policy settings within the tools themselves
Hard-blocked at the endpoint or network level to create a safety net against policy workarounds or configuration errors
Transitioned to managed access, where the tool isn’t completely blocked but instead routed through a dedicated corporate gateway that checks access permissions, and monitors usage patterns
Detecting AI systems
Spotting AI requires a multi-layered approach, as different detection methods complement each other and work best against specific types of AI.
Technology
What it can detect
DNS
Any AI tool with an identifiable domain
Web Gateway or NGFW
Any AI tool with a recognizable request-and-response fingerprint (API endpoint paths, domains, and other indicators). Web filters can inspect traffic content, and many gateways/NGFWs now feature a dedicated category for detecting and blocking generative AI
EPP/EDR
Locally deployed LLMs (running via Ollama, LM Studio, and similar shells), native desktop apps for ChatGPT or Claude, agentic browsers, and open-source AI agents. An indirect but strong red flag is the presence of Node.js, Python, Git, Docker, or other containerization tools on machines belonging to non-technical staff
Application control
Similar to EPP/EDR, this allows to immediately block unwanted applications right out of the gate
Browser control
AI-focused browser extensions and visits to AI-themed websites. This is a lifesaver if the corporate web gateway can’t inspect encrypted traffic
OAuth permissions requested by AI apps and services, as well as any third-party integrations plugging into core productivity hubs (Microsoft 365, Google Workspace, and others)
Naturally, almost all of these tools allow to do more than just spot AI — they let to block it entirely, or at the very least, sound the alarm for the team in charge.
Keeping an eye on OAuth
Popular office AI solutions — especially meeting assistants, email and calendar automation agents, and the like — gain access to corporate data by requesting OAuth permissions directly from communication, document workflow, or video conferencing platforms. If a user has the green light to grant these permissions to third-party apps, the resulting data leaks completely bypass the organization’s perimeter. Tools like EDR and NGFW won’t see a thing when a tool like Read.ai grabs recordings of every single meeting in, say, Microsoft Teams.
The most drastic — and often best — move is to block standard users from granting OAuth consent in the first place. Here’s how to handle the technical heavy lifting (Global Administrator, Application Administrator, or equivalent rights are needed):
Microsoft 365 / Entra ID
In the Microsoft Entra admin center, head over to Identity > Applications > Enterprise apps > Consent and permissions > User consent settings. There User consent for applications can be disabled (check out Microsoft’s full guide).
Google Workspace
In the Google Admin console, navigate to Security > Access and data control > API controls. Under Manage App Access, the trust level for all apps can be set: Trusted, Limited, Specific Google data, or Blocked. However, the real kicker here is the Unconfigured app settings subsection, which dictates what happens when a user tries to connect an unknown app. To seal this loophole, select Don’t allow users to access any third-party apps.
A separate subsection, Manage Google Services, permits fine-tuning exactly how third-party apps interact with Google Workspace and Google Cloud services. This allows to cut off access for each individual Google product (see Google’s official guide).
Salesforce
In Setup, use the Quick Find box to search for connected apps, then select Manage Connected Apps from the results. While settings are configured for each external app individually, all users can approve access by default. There isn’t a blanket block switch here; instead, Salesforce allows to opt for Admin approved users are pre-authorized (see the full Salesforce guide on this).
Slack
From the Admin settings menu, head to Apps and workflows -> App Management Settings. Tweak the Require approved apps setting by selecting Only allow pre-approved apps. Once that’s locked in, double-check that no rogue AI tools have slipped onto the approved list.
Have you ever tried to tally up how much you spend on subscriptions each month? Music, movies, gaming, language courses, delivery services, heated seats, and even the ability to chat with the Grok bot directly from your car — there’s a subscription for just about everything now. There’s even a subscription service specifically designed to… track your other subscriptions.
The number of subscriptions varies significantly depending on where you live, but statistically, 78% of adults worldwide have at least one paid subscription, with the average user juggling 5.6 active services. Furthermore, a large portion of these are family plans used by groups of close relatives… and sometimes other people: 37% of users share their subscriptions outside their immediate family.
Because subscription accounts, especially family plans, often contain sensitive personal data, they’ve become a prime target for cybercriminals. Today we look at how to manage your subscriptions securely, avoid having your accounts compromised, and keep from falling for scammers’ latest tricks.
Security of shared accounts and subscriptions
Why would anyone want to hack your subscription? Even if the service only offers entertainment, your account almost certainly contains sensitive information: your name, address, email, phone number, the names of other members, and other personally identifiable information. This data is then sold on the dark web and used for further attacks.
Attackers compromise subscription accounts either through social engineering and phishing, or by taking advantage of many users’ reliance on weak or leaked passwords. As we recently highlighted in our research, nearly half of all passwords worldwide can be cracked in less than a minute. Scammers then either resell existing subscriptions or slots in a family group at a discount, or they sign the victim up for new services, hoping the extra charges go unnoticed.
Finally, some middlemen don’t bother with hacking at all; they simply buy bulk subscriptions for a large number of devices, where the per-unit cost is typically much lower. They then resell individual slots in these plans on online marketplaces. As a result, a single “family” account can end up filled with people who are complete strangers to one another.
Sharing subscriptions with family and others
Many subscription owners think nothing of sharing access with family and friends. What could possibly go wrong?
The worst-case scenario from a security standpoint is when a single account is purchased and the owner shares the login and password with other users. This usually happens when people try to save money on a family plan by buying an individual subscription and sharing it. Some services even allow for different profiles, but they are all tied to a single account, meaning the credentials are shared. This is how streaming platforms like Hulu and Disney+ operate.
Sharing one account among multiple people significantly increases the risk of your credentials falling into the wrong hands. There’s no way to guarantee that everyone else is storing those details securely or that their devices aren’t infected with malware. Even without malware, it’s incredibly easy to accidentally hand over a password to attackers simply by signing in to the subscription service over unprotected public Wi-Fi.
It’s entirely possible that the password you kindly shared with some friends has already surfaced in some corner of the dark web, and you may soon lose access to your account. Furthermore, if you reuse the same password across different sites and apps, your other accounts are now in the crosshairs as well.
The second scenario is when each group member has an individual account. Many services now allow you to add extra users to a subscription at no additional cost, and most owners are happy to give away these free slots. Even then, you shouldn’t let your guard down: a breach of just one of these accounts can still leak sensitive information, such as family members’ names, addresses, billing info, and other subscription-related data.
How to protect your subscriptions (and your wallet)
To keep your and your loved ones’ personal data private and your accounts under your control, follow these simple rules.
Use strong account security
To do this, learn — and teach your friends and family — how to use password managers, two-factor authentication, or passkeys.
If you and your loved ones rely on memory to store passwords, there’s a high probability that you’re reusing the same one across multiple services. This is a major blunder: data breaches happen all the time, and a single compromised password gives attackers access to your other accounts.
The simplest solution is to use a password manager that generates and remembers complex, unique passwords for every site and service on your behalf. All you have to do is remember the single main password for its encrypted vault. Additionally, Kaspersky Password Manager doesn’t just store and create passwords; it can also check if they’ve appeared in leaked databases, and sync your credentials across all your devices.
Additionally, a password manager provides a robust defense against phishing: unlike a human, who can easily be misled by a sign-in form that looks almost identical to the real thing and is hosted on a look-alike domain, a password manager won’t fall for the trick. It’ll only offer to autofill your saved login and password on the specific site or service for which they were originally stored.
Two-factor authentication (2FA) is an extra layer of verification the system requests after you enter your password — such as an SMS code or a one-time code from an authenticator app. Whenever technically possible, be sure to enable 2FA on every account linked to a subscription. This applies to the subscription services themselves, as well as any third-party accounts you use to sign in, such as Google, Apple, or Facebook.
We recommend storing your two-factor authentication tokens and generating the one-time codes — which refresh every 30 seconds — inside Kaspersky Password Manager. This significantly lowers the chances of someone hijacking your account. Even if an attacker somehow discovers or guesses your password, they won’t be able to get the code without physical access to your device.
Finally, you can ditch passwords (almost) entirely by switching to passkeys. We’ve previously covered what this password alternative looks like and the specifics of using it. Currently, this is the most breach-resistant authentication system out there. Its main drawback has been the difficulty of syncing passkeys across different ecosystems, like Windows and iOS, but the updated version of Kaspersky Password Manager can now save and sync passkeys across Windows, macOS, iOS, and Android devices, making that issue a thing of the past.
Don’t overlook device security
Even a complex password and 2FA aren’t reasons to let your guard down. An attacker can infect your device with an infostealer: malware designed to swipe things like session cookies from your browser, app configuration files, and other sensitive data. Session cookies allow you to stay signed in without re-entering your credentials every time; however, if scammers get their hands on them, they can sign in to the service as you — even without knowing your username or password. This makes a proactive approach essential, especially if you use Chrome, Edge, Opera, or other Chromium-based browsers on Windows. We recommend installing Kaspersky Premium on all your devices; it includes Kaspersky Password Manager in addition to comprehensive protection against cyberthreats.
Only share subscriptions with people you trust
Otherwise, you might be asking for trouble. For example, if you share a Steam subscription with a friend who cheats, both of your accounts could end up banned. Furthermore, never try to let someone else into your personal account or individual subscription. Sharing your password with others is usually a violation of the terms of service, and can result in your account being blocked.
Make sure there are no strangers in your family group
To do this, periodically check active devices and sessions in your subscription settings. If you see an unrecognized device in the authorized list, terminate that session — or all of them — and change your account password immediately. Signing back in on a few devices is much easier than trying to recover a hijacked account.
And remember: don’t let your own habits compromise your security. If you’re visiting friends, on vacation, or on a business trip and use a local computer or smart TV — or if you sign in to your account from a public computer — don’t forget to sign out when you’re done. Otherwise, the next person to use that device might find themselves with free subscriptions or, even worse, access to your email or cloud photo stream.
Don’t take the bait
Watch out for phishing emails and messages spoofing legitimate services. If you receive a notification about a “need to update your billing details”, or a claim that a “new user has been added” to your family plan, don’t rush to click any links or open attachments. Links can lead to a phishing page, and attachments may hide malware. Scammers often use email addresses and domains that look nearly identical to the real ones — for instance, by swapping l (lowercase L) for I (uppercase i), or using a familiar name in a different domain zone.
Unfortunately, phishing pages are often indistinguishable from the originals now that AI is being used for high-quality design and layout. Since spotting every red flag yourself is increasingly difficult, it’s best to delegate anti-phishing protection to Kaspersky Premium. It will alert you to suspicious sites, saving your money and keeping your peace of mind.
Lastly, some scammers lure users in with freebies like fake gift subscriptions for Telegram Premium. The victim is asked to visit a phishing page mimicking the Telegram login screen and sign in to their account to claim the gift. The result isn’t hard to guess: instead of a premium subscription — a hijacked account. Recently, scammers have even learned to use mini-apps to steal credentials directly inside Telegram under various pretexts — ranging from gift giveaways to claims that you must move to a new chat because the old one was blocked.
Avoid buying subscriptions from third-party sellers
You can often find subscription offers on marketplaces and retail platforms at prices significantly lower than what the official provider charges. More likely than not, that tempting price hides a hacked account or a family group that you could be kicked out of at any moment, because the family admin is either the seller or a random user. Furthermore, sharing a family plan with strangers from around the world is a violation of terms for many services.
How to get rid of unwanted subscriptions
Now that we’ve covered subscription security, what about those extra subscriptions that quietly eat away at your balance every month? Research shows that users typically underestimate how many active subscriptions they have and how much they spend on them; they also frequently forget to cancel auto-renewals for subscriptions they no longer use, or auto-charges after the trial period ends.
If you suspect you’re in that boat, start your investigation with your own bank statements. Recurring charges for the same amount can be a subscription you’ve forgotten about. Check who received the payment; if the name doesn’t ring a bell, do an online search on the company. It’s also worth searching your email box for the merchant name or the payment amount; this can help you track down subscription notifications and figure out what exactly you’re paying for. And don’t forget to check your spam folder, as that’s where subscription alerts often end up.
Now, let’s look at how to check and cancel active subscriptions purchased through the App Store and Google Play.
For Android users
Open Settings on your device.
Tap Google, then tap your profile picture, and go to Google Account.
Go to Wallet & subscriptions.
If you’re the family group manager, you’ll be able to see the purchase history for other family members.
For iOS users
Open Settings on your device.
Tap your profile picture at the top of the menu.
Go to Subscriptions.
Note: to manage your iCloud subscription, you’ll need to go to the specific iCloud section located just below Subscriptions. In the Family Sharing section, if you’re the one who set it up, you can view the subscription and purchase history for all family members.
Among the various tools in the Kaspersky portfolio is a dedicated platform for securing containerized environments. But in this post, I want to talk about Kaspersky Container Security (KCS) — not as a vendor representative, but rather as a member of a team that actively uses this solution in their daily work. Our Product Security Team is responsible for establishing secure development processes across the company. We’re involved in every stage of the software development life cycle, and our priority is helping product teams catch security issues early so they can stay on schedule for their releases. To achieve this, we’ve built several workflows, one of which focuses specifically on container security. That’s exactly where we lean on our own Kaspersky Container Security platform.
Container security solutions are typically viewed first and foremost as image scanners for the container registry. However, Kaspersky Container Security (KCS) is more of a comprehensive security platform for container environments that handles multiple tasks by virtue of its end-to-end integration into the container workflow. While it certainly includes a container image scanning scenario — which is undeniably important — our experience with KCS has shown that its real value becomes apparent when it’s integrated into several points along the workflow at once:
Regular builds
Artifact verification prior to release or deployment
Monitoring of containers already running in the cluster
The baseline scenario: how KCS scans images
At its core, the process is a standard one. KCS checks images for typical container issues: known vulnerabilities, malware, hardcoded secrets, and misconfigurations. However, the scan result isn’t just a single, abstract verdict. The system calculates a risk rating based on the findings, providing a clear picture of the asset’s security posture. In practice, this is incredibly useful because teams don’t just see a “bad image” message; they get a transparent breakdown of exactly what’s driving the risk and what needs to be fixed first.
But that’s not all. KCS works well for scenarios where it’s not enough to just find a problem — you need to tie it to the artifact’s life cycle. When a team is managing hundreds of builds, periodic registry scanning isn’t enough, and it almost always requires manual intervention. You need to know which pipeline introduced the risk, which policies were triggered, and what the next steps are. KCS provides this essential link.
Advanced scenario: CI/CD integration
One lesser-known KCS feature is its full-scale scanning capability within CI/CD pipelines. For our team, this is the most effective way to use KCS. The logic is straightforward: you integrate the scanner into the pipeline, and the scan results appear directly in the execution logs. They’re also sent to the solution’s central console, where they’re logged in a dedicated CI/CD section that links the findings to the artifact name, scan time, pipeline, and severity level.
In a CI/CD environment, you can scan images from tar-archives or directly from Git repositories. Out of the box, it supports GitLab, Jenkins, TeamCity, and GitHub Actions; in practice, KCS can be integrated into any pipeline orchestrator.
Another critical aspect of using KCS in CI/CD involves security policies. Our solution uses a model where policies allow for not just collecting results, but also controlling the behavior of the pipeline itself. This comes in handy for phased rollouts. You can start in audit mode, and then gradually move toward failing builds when secrets, critical misconfigurations, or vulnerabilities are detected. This evolutionary approach generally works better than simply flipping a switch to block it all at once.
How KCS helps in our workflows
We run our own composition analysis system, so we don’t treat KCS as a single source of truth. Instead, it serves as a powerful extra layer in our workflows, and that’s exactly where we find the most value.
While our in-house composition analysis system handles component tracking, dependencies, and code-level risk assessment, KCS excels at securing the container perimeter. It takes care of technical image scanning and CI/CD security, while aggregating reports on container artifacts. It doesn’t conflict with our internal analysis; it reinforces it right where containers receive actual workloads.
This is particularly useful for us in two scenarios. First, it provides early-stage artifact control during development. Second, it acts as a gatekeeper during release acceptance. We no longer debate risks sometime after the release; we catch them at the exact point where the team can still quickly fix a Dockerfile, Helm chart, or config set without a lengthy approval chain.
The way it handles a software bill of materials (SBOM) is also noteworthy. Our system relies primarily on up-to-date, relevant SBOMs. KCS offers modes specifically for processing SBOMs, and can even output scan results in that same format. In this regard, KCS integrates seamlessly with our internal processes, allowing us to fit it into our existing workflows rather than the other way around.
Why KCS is more than just a scanner to us
Its other powerful layer is cluster security. At this stage, KCS evolves beyond being just an image-scanning tool. It features runtime policies for containers and nodes, audit and blocking modes, and a set of security profiles. In practical terms, this means KCS can be used not only to find vulnerabilities within an image, but also to monitor what the container is actually doing once it’s live. Policies can account for image provenance, digital signatures, restrictions on capabilities and volumes, and even the processes and network connections running inside the container.
When a problem is detected, you have the option to log the results in audit mode first rather than blocking the process immediately. In production environments, this is always the smarter move. Another vital tool is ensuring trusted image provenance. KCS supports digital signature verification, which shifts the focus from simply finding CVEs to securing the company’s entire software supply chain.
Reporting capabilities
KCS does more than just display the issues it detects; it serves as a comprehensive reporting source. It can generate reports on images, accepted risks and Kubernetes benchmarks.
Generated reports are available in HTML, PDF, CSV, JSON and XML formats, with specific support for SARIF for detailed reporting — which is ideal for integrating into AppSec workflows. As for the SBOMs mentioned above, the scanning scenarios can output artifacts and results in CycloneDX and SPDX formats, making it easy to plug into existing processes.
Why we continue to use KCS
To put it simply, KCS complements our workflows perfectly — not because it solves every single problem, but because it integrates so effectively into engineering scenarios.
We also appreciate that the product team listens to our feedback. The KCS team actually incorporates our practical operational requests into their development roadmap. For example, deep SBOM integration and specific report types were added to KCS as a direct result of our hands-on experience.
To sum it up, when integrated correctly, Kaspersky Container Security helps cover several areas at once: from basic container scanning, to CI/CD and cluster security. In our experience, it provides real value within a live container ecosystem. You can learn more about the solution on the official KCS page.
AI security covers more than just data theft prevention, restricting rogue AI agents, or stopping assistants from giving harmful advice. A relatively simple but rapidly scaling threat has emerged: attempts to hijack computational power and exploit someone else’s neural network for personal gain. This is known as LLMjacking. With AI compute costs widely predicted to surge dramatically, the number of attackers driven by these motives is poised to grow. Consequently, when deploying proprietary AI servers and their supporting ecosystems like RAG or MCP, it’s critical to establish rigorous security measures from day one.
Statistics from a honeypot
The speed and scale of these resource-hijacking attempts are best illustrated by an experiment documented in detail in April 2026. The investigator configured a Raspberry Pi to masquerade as a high-performance private AI server, and made it accessible from the internet. When queried, it reported the availability of Ollama, LM Studio, AutoGPT, LangServe, and text-gen-webui servers — all tools commonly used as wrappers for locally hosted AI models. The server also appeared ready to accept API requests in the OpenAI format, which has become the industry standard.
All these services were seemingly powered by a local instance of Qwen3-Coder 30B Heretic, one of the most powerful open-source models, with its safety alignment removed. To throw in a sweetener, the honeypot reported the presence of various RAG databases and an MCP server with tempting capabilities like get_credentials on board.
In reality, the Raspberry Pi was simply hosting 500 pre-saved responses from an actual Qwen3 model, with a lightweight script selecting the most relevant answer for each incoming query. This setup was enough to pass a superficial check while allowing the researcher to probe the attackers’ intentions.
According to the author, Shodan, a popular internet scanning service, discovered the server within three hours of its going live. Just one hour later, requests resembling capability reconnaissance began pouring in. Over the following month, the server handled more than 113 000 requests from thousands of unique IPs, with 23% of that traffic specifically targeted at discovering AI capabilities and exploiting local LLMs and AI agents.
Requests to endpoints like /api/tags and /v1/models allow attackers to fingerprint which models are hosted on a server, while scanning for /.cursor/rules typically precedes an attempt to exploit an AI agent. Similarly, checking /.well-known/mcp.json serves as an inventory of the victim’s MCP servers. While the author makes no mention of the total number of attacks that progressed beyond simple scanning, there were 175 active attempts to hijack the LLM during the final week of the experiment alone.
What are the attackers after?
Based on the researcher’s observations, none of those targeting the decoy server attempted to execute arbitrary code or gain root access. (Editorial note: this is surprising and may point to gaps in logging.) Almost all attacks were aimed at siphoning resources. For example, the following activities were logged during the experiment:
A well-structured attempt to parse technical documentation for a microprocessor
A prompt to write an erotic novel
Requests to parse and structure social media text data regarding new vulnerabilities
An attempt to call Anthropic models using the compromised server as an API proxy
It’s worth noting that the reconnaissance of AI resources uses standardized and rapidly evolving tools. Requests from an application named LLM-Scanner originated from the infrastructure of seven different cloud providers across eight countries, suggesting that the raiders have put established methodologies in place, as well as specialized platforms for sharing techniques. By the third week of the experiment, the scanner had been updated with an additional check: it now used simple abstract questions to determine whether it’s interacting with live AI or a honeypot returning canned responses.
Among the non-specific attacks, the experiment recorded numerous attempts to exfiltrate credentials from the .env file. Attackers systematically hunted for this file across every conceivable directory on the server. Leaving an .env file publicly accessible is one of the most elementary mistakes when deploying projects on Laravel, Node.js, and other frameworks, yet it remains a common oversight — particularly among beginners and vibe coders. Consequently, attackers have every reason to expect their efforts to pay off.
Conclusions and defense tips
Scanning publicly accessible servers and attempting to exploit them is nothing new, but the rise of LLMs gives attackers another way to monetize their efforts — one that’s both highly lucrative for them and devastating for their victims. To understand how massive these attacks could become, look at their closest counterpart: the cryptojacking market — where criminals mine cryptocurrency using stolen computational resources. That market grew by 20% in 2025 alone. As AI-powered solutions proliferate, and as major providers hike subscription costs while local AI chips remain in short supply, we should expect LLMjacking to become an industrial-scale phenomenon.
Key defensive measures for private AI infrastructure
For AI systems running locally on a single machine, ensure that servers like LM Studio, Ollama, or similar are configured to accept connections only on the local interface (localhost), rather than all available network interfaces. This restricts LLM access to the host machine itself, and prevents the AI from being reachable over the internet.
For servers handling remote requests — even if the server only operates within a local corporate network — implement robust authentication and authorization rather than relying solely on API key validation. Solutions based on OIDC or OAuth2 with short-lived tokens are the most effective. This not only defends against LLMjacking, but also allows for more granular tracking of user activity, and prevents API key abuse. Furthermore, keys must be protected from more than just external attackers; a growing risk is the misuse of keys by AI agents themselves. This applies to LLM interfaces as well as MCP, RAG, and others.
Use network segmentation and IP allowlists to give AI server access only to the departments, employees, and services that require it.
Ensure that all client-server connections are secured with a current version of TLS.
Apply the principle of least privilege by separating access to specific services; for instance, MCP and LLM components should have their own distinct access tokens.
Ensure an EDR security agent is installed on all workstations and servers, including those hosting AI models.
Monitor AI resource consumption, establish usage quotas for different employee roles, and set up alerts for anomalous activity spikes.
Maintain detailed logs of LLM responses and requests made to the model and its supporting tools. Integrate these data sources with your SIEM. Ensure logs are resilient against tampering or deletion.
To put it simply, the classic logic of a SIEM system works as follows: if event A occurs, followed by event B, this may be a sign of an attack, and an information security specialist should be notified. But in today’s environment, this simple scenario is increasingly failing. Just recently, our experts analyzed a high-profile incident: attackers compromised the update infrastructure of the popular Notepad++ software, and distributed malware via the update mechanism. It’s simply impossible to have rules in place in advance that are specifically designed to counter such scenarios.
The attacks themselves have become more sophisticated: attackers use legitimate tools, they attack through the supply chain by compromising software outside the corporate perimeter, stretch out their scenarios over time, and disguise their actions as normal activity. In other words, they do not “break into” the infrastructure; more often than not, they log in and use legitimate software. As a result, the classic fixed rules of the past either fail to trigger, or generate too many false alerts. This is what prompted the shift toward more flexible correlation scenarios.
Dynamically updated SIEM content
Correlation content today isn’t a static set of rules, but a process: it’s constantly evolving and adapting to current threats. In 2025 alone, we released 55 rule-package updates for different versions and languages of our Kaspersky SIEM system. In just one year, we added 10 new rule packs, as well as 250 detection rules and numerous improvements to existing content. This year, we’ve already added 43 new rules and refined another 63. In total, this amounts to over 850 rules covering a significant portion of the MITRE ATT&CK framework.
Kaspersky SIEM rules are written based on insights from our experts who analyze real-world, recent attacks: we primarily draw on the findings of our managed detection and response (MDR) service and our threat research. As a result, our rules cover scenarios — from reconnaissance to privilege escalation — that involve the latest approaches used by attackers. For example, we detect the use of new attack techniques such as ToolShell.
In addition to scheduled updates, the team regularly releases so-called emergency content — rule sets for rapid response to new and unexpected attack techniques. In February, for example, detection rules were released for authentication bypass in Fortinet products via the SSO mechanism: attackers used specially crafted SAML requests to gain access to systems without credentials.
From events to attack chains
Moreover, modern SIEM rules no longer describe individual events, but rather sequences of actions. Scenarios are built around the stages of an attack: from initial access, to privilege escalation and persistence. Kaspersky SIEM’s effectiveness is enhanced through integration with Kaspersky EDR and dedicated rule sets for Active Directory, which implement dozens of attack detection scenarios at various stages. This approach allows us to see not just individual signals, but the full picture.
Integration and internal visibility
Another way to improve the effectiveness of an SIEM system is to expand data sources. A classic SIEM aggregates events from different levels of the infrastructure: from logs to telemetry from endpoints and internal systems. In addition to this, our SIEM system includes specialized rule sets for our other solutions (Kaspersky Security Center, Kaspersky Security for Mail Groups, K Anti-Targeted Attack platform), which allow monitoring of administrator actions, authentication, and service status. As a result, the system becomes a tool not only for detecting attacks, but also for monitoring internal activity.
Overall, SIEM is no longer just a set of rules, but has evolved into a continuously updated detection system. Its effectiveness is determined not by the number of detections, but by their relevance, coherence, and how accurately they reflect the actual actions of attackers. Stay up to date regarding our Kaspersky Unified Monitoring and Analysis Platform (SIEM) on its official product page.
Every year, hundreds of millions of real user passwords leak onto the dark web. We analyzed 231 million unique passwords from dark-web leaks between 2023 and 2026, and the conclusions are bleak: the vast majority are extremely weak. To crack 60% of these passwords, a hacker needs only an hour and a few dollars in their pocket. Furthermore, password cracking is accelerating by the year; in our similar 2024 study, the percentage of vulnerable passwords was lower.
Today we’re looking at just how reliable the average password is (spoiler: not really), and how you can secure your data and accounts using more robust methods. At the same time, we’ll highlight the patterns most commonly found in actual user passwords.
These days, passwords are almost never stored in plain text. For instance, if you create an account with the password “Password123!”, the server won’t store it as-is. Instead, the password is hashed using specific algorithms, turning it into a fixed-length string of letters and numbers (a hash) which is what actually stays on the server. For example, here’s what the MD5 hash for “Password123!” looks like:
2c103f2c4ed1e59c0b4e2e01821770fa.
Every time the user enters their password, it’s converted into a hash and compared against the one stored on the server; if the hashes match, the password is correct. If an attacker gets their hands on this hash, they have to decrypt it to recover the original password — this is what’s known as “password cracking”. This is typically done using owned or rented GPUs, and several methods can be employed for the crack:
Exhaustive enumeration (brute force). The computer tries every possible combination of characters, calculating the hash for each one. This method is the easiest way to crack short passwords, or those consisting of a single character set (such as digits only).
Rainbow tables. A total nightmare for anyone with a simple password, this is essentially a “phone book” for passwords whose hashes have already been cracked via brute force or smart algorithms. All an attacker has to do is find a matching hash and see which password corresponds to it.
Smart cracking. These algorithms are trained on databases of leaked passwords. They understand the frequency of different character combinations, and run their checks from the most likely to the least popular sequences. They account for dictionary words, character substitutions (a → @ or s → $), and consider common password structures like “dictionary word + number + special character”, while checking hashes against rainbow tables. Combining these methods significantly accelerates the cracking process.
Beyond that, attackers can also intercept passwords in plain text. There are numerous ways to do this, ranging from phishing (where a victim is lured to a fake web page and enters their password voluntarily) and keyloggers that capture keystrokes, to stealers or Trojans that swipe documents, cookies, clipboard data, and more. Unfortunately, many users keep their passwords as plain text in notes, messaging apps, and documents, or save them in browsers where attackers can extract them in seconds.
Every year, we track around a hundred million plain-text password leaks. We use these databases to warn Kaspersky Password Manager users if their data has been compromised. To address the most frequent question we get on this: no, we don’t know our users’ passwords. We’ve explained in non-techie language exactly how we compare your passwords to leaked ones without actually knowing them — and why neither your passwords stored in Kaspersky Password Managernor even their hashes ever leave your device — in our overviews of our leak analysis technology and our password manager’s internal architecture. Give them a read; you’ll be surprised by just how elegant the design is.
60% of passwords are cracked in under an hour
We expanded the database from our previous study by an additional 38 million real passwords posted by attackers on dark-web forums and compared the results. Testing was conducted using a single RTX 5090 GPU for passwords hashed with the MD5 algorithm. The data for the analysis was obtained from our Digital Footprint Intelligence service. You can review the algorithm we used to assess password strength in our article on Securelist.
Unfortunately, passwords remain as weak as ever, while cracking them becomes faster and easier with every year. Today, 60% of passwords can be cracked in less than an hour; two years ago, that figure was 59%. But the truly frightening part is something else: nearly half of all passwords (48%) are cracked in less than a minute!
Cracking time
Percentage of passwords crackable within this time in 2024
Percentage of passwords crackable within this time today
Less than a minute
45%
48%
Less than an hour
59% (+14%)
60% (+12%)
Less than 24 hours
67% (+8%)
68% (+8%)
Less than a month
73% (+6%)
74% (+6%)
Less than a year
77% (+4%)
77% (+3%)
More than a year
23%
23%
Password cracking time: two years ago and today
Attackers owe this boost in speed to graphics processors, which grow more powerful every year. While an RTX 4090 in 2024 could brute-force MD5 hashes at a rate of 164 gigahashes (billion hashes) per second, the new RTX 5090 has increased that speed by 34% — reaching 220 gigahashes per second.
And although a high-end video card like that currently retails for several thousand dollars, the price tag isn’t much of a barrier: there are plenty of cheap cloud services available for renting GPU computing power. Depending on the configuration and the model, rental costs range from a few cents to a few dollars per hour. As we’ve seen, one hour is all an attacker needs to crack three out of every five passwords they’ve found in a leak. Plus, depending on the scale of the task, they can always rent ten or even a hundred GPUs instead of just one…
It’s worth noting that cracking every password in a dataset doesn’t take much longer than cracking a single one. During each iteration, once the attacker calculates a hash for a specific character combination, they check if that same hash exists anywhere in the dataset — and the larger the dataset, the easier it is to find a match. If a match is found, the corresponding password is flagged as “cracked”, and the algorithm moves along to the next one.
Which passwords are vulnerable?
The strength of any password depends on its length, content variety, and the randomness of that content. Passwords created by humans turn out to be the least resilient — unfortunately, humans are quite predictable. We use dictionary words and character combinations that smart algorithms have long since mastered, we avoid long random strings, and patterns can be found even in keystrokes we believe are random. Interestingly enough, passwords generated by AI still carry the fingerprints of a human approach; we covered this in a separate post on how to create a strong yet memorable password.
Password length is the primary factor affecting cracking time. As you can see from the table below, it takes less than 24 hours to crack almost any eight-character password.
Percentage of varying password lengths crackable within a given timeframe
But the predictability of your password is just as important. Think you’re boosting security by adding a number or a special character to a memorable word? You are, but only slightly. The patterns people use to create passwords are easily predictable and, at times, pretty amusing — though this is no laughing matter.
What we learned about password patterns
Analysis of over 200 million passwords revealed characteristic patterns that allow smart algorithms to crack user passwords with ease.
Pick a number
More than half of all passwords (53%) end with one or more digits, while nearly one in six (17%) starts with a number. Every eighth password (12%) contains sequences that look a lot like years — ranging from 1950 to 2030 — and one in ten (10%) specifically falls between 1990 and 2026. This most likely happens because folks add their birth year (or that of someone close), some other significant year, or the year they created the password or account. Fun fact: based on the distribution of these dates, it suggests that the most active internet users were born between 2000 and 2012.
However, among all numeric combinations, the most popular turned out to be… you guessed it: “1234”. Overall, patterns involving sequential keyboard presses (“qwerty, ,”ytrewq”, and the like) appear in 3% of passwords.
Special characters aren’t a silver bullet
Most password policies in recent years require at least one special character. The absolute winner in this category is the @ symbol: it appears in one out of every 10 passwords. The period (.) comes in second, followed by the exclamation point (!) in third.
Love rules the world… and Skibidi Toilet does too
Emotionally charged words often form the foundation of a password, and despite everything, positive words are more common. Frequently occurring examples include “love”, “angel”, “team”, “mate”, “life”, and “star”. That said, negativity pops up too — mostly in the form of common English swear words.
Interestingly, viral memes are reflected in passwords as well. Between 2023 and 2026, the use of the word Skibidi in passwords skyrocketed 36-fold! Naturally (see the link if it doesn’t seem natural), “toilet” saw a boost too, though to a lesser extent.
Users tend to keep their passwords unchanged for years
More than half of the passwords (54%) we identified in recent leaks have surfaced before. Part of this can be explained by the same data migrating from one dataset to another. However, there’s a much more troubling reason too: many users simply haven’t changed their passwords in years.
Analyzing the dates found within passwords shows that combinations containing the years from 2020 through 2024 remain popular. It seems people add the current year to their password when they create it — and then forget about it for several years. This actually allows us to calculate the average lifespan of a password: about three to five years.
This is a dangerous trend. For one, smart algorithms can crack much more complex passwords over that kind of timeframe. Secondly, the longer your password remains unchanged, the higher the probability it will leak — whether through a breach, malware infection, or a phishing attack.
The situation gets even worse when the same password is used across multiple accounts. In this case, attackers don’t even need to crack anything; they just need to find your password in a single leak and plug it into other sites.
How to protect your passwords and accounts
If you’ve realized while reading this post that your own passwords are among those easily crackable — don’t panic. We’ve put together a list of simple but essential tips for you.
Use a password manager
The weakest passwords are the ones people come up with themselves. Creating and memorizing hundreds of sequences of 16–20 random characters (since every site requires a unique, long password) is a daunting, unrealistic task.
That’s why you should delegate password generation and storage to our password manager. It doesn’t just create and store complex, randomized passwords in an encrypted format; it also syncs them across all your devices. To decrypt your vault, you only need to remember one main password that no one knows but you — our guide on mnemonic passwords can help you with that.
Don’t store passwords as plain text
Whatever you do, never write down passwords in files, messages, or documents. They lack the robust encryption provided by a password manager. Furthermore, these kinds of notes fall into the hands of attackers instantly if you happen to pick up a Trojan or an infostealer.
Don’t store passwords in your browser
Many users save their passwords in their browsers — especially since they conveniently offer to do it automatically. Unfortunately, research shows that malware has evolved to extract these passwords from all popular browsers almost instantly. Kaspersky Password Manager can help you import saved passwords from your favorite browser — just follow our simple, three-step guide. Most importantly, don’t forget to clear the browser’s password storage once the import is complete.
Switch to passkeys
Wherever possible, use passkeys — a cryptographic replacement for passwords. In this setup, the service stores a public key, while the private key remains on your device and is never transmitted. During login, the device simply signs a one-time request. Additionally, passkeys are tied to a specific domain, meaning phishing attacks using spoofed addresses won’t work. Kaspersky Password Manager allows you to store both passwords and passkeys, solving the problem of syncing them across different ecosystems, including Windows, Android, macOS, and iOS.
Set up two-factor authentication
Enable two-factor authentication wherever possible. Even if your password is compromised, a properly configured 2FA setup makes it extremely difficult for the attacker to access your account. For maximum security, skip the one-time codes sent via SMS and use authenticator apps instead — and yes, Kaspersky Password Manager comes in handy here, too.
Practice good digital hygiene
Remember, storing your passwords correctly is only half the battle. It’s crucial to follow the rules of digital hygiene: avoid downloading unverified files, pirated software, cheats, or cracks, and don’t click on random links. The number of infostealer attacks has been steadily rising in recent years, which means you need a robust security solution for full protection. We recommend Kaspersky Premium — it protects all your devices from Trojans, phishing, and other threats. Besides, the subscription includes our password manager.
For those serious about account security, check out our collection of posts on passwords, passkeys, and two-factor authentication:
Malicious actors have developed a new way to steal data stored by Chrome for Windows. Researchers discovered the technique while analyzing a fresh build of an infostealer known as VoidStealer. The new method allows the malware to bypass Chrome’s Application-Bound (App-Bound) Encryption (ABE), a mechanism intended to protect session cookies and other valuable information stored in the browser.
Google hoped this mechanism would secure the master key Chrome uses to encrypt all sensitive data. Unfortunately, this isn’t the first time malware authors have found a workaround for this defense — leaving secrets stored in Chrome vulnerable once again.
How App-Bound Encryption works in Chrome
Google introduced App-Bound Encryption in July 2024 with the release of Chrome version 127. The company’s announcement mentioned infostealers snatching cookies from Chrome users on Windows as the primary problem ABE was intended to solve. We’ve already covered in detail what these files are and the consequences of their theft, so we’ll only briefly recap the main facts here.
Cookies are small files that the browser saves to the user’s device at a website’s request to remember various site settings. Of particular value to attackers are session cookies, which are used for automatic authentication on websites. It’s thanks to these files that we don’t have to enter a username and password every time we revisit a site.
But this convenience carries a risk: stealing these files allows an attacker to use an already-authenticated session without entering a username or password. This allows them to impersonate the user, which can lead to account hijacking, theft of personal or financial data, and other adverse consequences.
Infostealer Trojans are particularly dangerous for Chrome users on Windows. This is because, on this OS, Chrome previously relied solely on the standard built-in Data Protection API (DPAPI). With this system encryption mechanism, applications don’t need to create and store encryption keys to protect data.
The limitation of DPAPI is that it doesn’t protect data from malware that’s already successfully compromised the system and is capable of executing code on behalf of the logged-in user. This is exactly what stealers exploit: since they typically run with the user’s privileges, they can simply request DPAPI to decrypt the browser’s protected data.
The ABE mechanism was designed to solve that specific problem. The core idea is right in the name: App-Bound Encryption means the encryption is tied to a specific application. To achieve this, a separate service running with system privileges is responsible for protecting the key used to encrypt Chrome’s data. It verifies which application is requesting access to the key, and denies the request if it doesn’t originate from Chrome.
Chrome’s App-Bound Encryption (ABE) was designed so that only Chrome itself could retrieve the master key needed to decrypt the browser’s stored data. Source
As a result, the architects of this feature assumed that to access ABE-protected browser data, an infostealer would either need to escalate its privileges to system-level, or inject malicious code directly into Chrome. In theory, this should have made attacking Chrome significantly harder and reduced the effectiveness of mass-market infostealers. As you might have guessed, things didn’t go quite that smoothly in practice.
Previous successful bypasses of Chrome’s ABE
Just a couple of months after Google announced the implementation of App-Bound Encryption in Chrome, many infostealer developers claimed they’d already bypassed the protection. Among them were the creators of Meduza Stealer, Whitesnake, Lumma Stealer, and Lumar (also known as PovertyStealer).
Lumma stealer developers announce a bypass for Chrome’s App-Bound Encryption in a new version of the malware
Of course, you shouldn’t take malware developers at their word, but legitimate security researchers were able to confirm at least some of the claims. Bypasses for Google Chrome’s new data protection feature did become available almost immediately after its release.
A month later, in October 2024, tech enthusiast Alex Hagenah published a tool on GitHub called Chrome-App-Bound-Encryption-Decryption to bypass Google’s new security mechanism. Analysis of the tool’s code revealed that its author used roughly the same methods that attackers were already heavily exploiting.
What followed was a game of cat and mouse: security researchers and stealer developers came up with new tricks to circumvent App-Bound Encryption, while Google patched the newly discovered loopholes with varying degrees of success.
VoidStealer — a new data-nabbing menace
This brings us to recent events: in March 2026, news broke about a stealer named VoidStealer, which utilizes a brand-new and, by all accounts, highly effective method for bypassing ABE.
VoidStealer developers advertising a new method for bypassing ABE. Source
The malware authors developed an attack technique that targets the brief moment when the master key sits in the browser’s memory in plaintext. This occurs because, at a certain point, the browser inevitably has to decrypt its data to actually use it — for instance, to automatically sign in to a website with the relevant session cookie or to access saved credentials.
To exploit this window of opportunity, the malware attaches itself to the Chrome process as a debugger — a tool that allows one to control a program’s execution, pause it, and inspect its memory. In legitimate scenarios, these tools are used by developers to find and fix bugs, analyze application behavior, and test performance.
The malware identifies the specific section of code where data decryption takes place. It then sets a breakpoint at that location; when the program’s execution reaches that point, the browser effectively freezes. This is how the malware catches the exact moment the master key is sitting in RAM in plaintext; it then reads the key directly from memory.
It’s worth noting that everything mentioned above also applies to other Chromium-based browsers that use ABE, including Microsoft Edge, Brave, Opera, Vivaldi, and others.
How to avoid falling victim to infostealers
The scale of VoidStealer’s reach could be significant, as its developers operate under the malware-as-a-service (MaaS) model. This means they rent out the ready-made tool to other attackers, so they don’t need to develop custom malware from scratch.
This situation demonstrates that relying solely on built-in security mechanisms isn’t enough. Unfortunately, stealer developers are coming up with new workarounds faster than browser and operating system developers can roll out patches.
Here’s what users can do about it:
Avoid installing programs from suspicious sources. This will minimize the chances of malware infiltrating your system.
Learn how ClickFix attacks Lately, stealers have frequently been distributed using this specific malicious tactic.
Keep your OS and software updated on all devices. Timely updates help patch many of the vulnerabilities that malware exploits.
Install a robust security solution on all your devices. It’ll block suspicious activity in real time and alert you to potential threats.
As an added precaution, avoid storing passwords and bank card info in Google Chrome or your Notes app, as these are the first places any self-respecting stealer looks. Instead, use a secure password manager.
Stealers are hunting for your data, finding ways to infiltrate both computers and smartphones alike. To protect yourself from theft, check out our other related posts:
Our experts have discovered a large-scale supply chain attack via DAEMON Tools – software for emulating optical drives. The attackers managed to inject malicious code into the software installers, and all trojanized executable files are signed with a valid digital signature of AVB Disc Soft – the developer of DAEMON Tools. The malicious version of the program has been circulating since April 8, 2026. At the time of writing, the attack is still ongoing. Researchers at Kaspersky believe this is a targeted attack.
What are the risks of installing the malicious version of DAEMON Tools?
After the Trojanized software is installed on the victim’s computer, a malicious file is launched every time the system starts up – sending a request to a command-and-control server. In response, the server may send a command to download and execute additional malicious payloads.
First, the attackers deploy an information gatherer that collects the MAC address, hostname, DNS domain name, lists of running processes and installed software, and language settings. The malware then sends this information to the command-and-control server.
In some cases, in response to the collected information, the command server sends a minimalistic backdoor to the victim’s machine. It’s capable of downloading additional malicious payloads, executing shell commands, and running shellcode modules in memory.
The backdoor can be used to deploy a more sophisticated implant dubbed as QUIC RAT. It supports multiple communication protocols with the command-and-control server, and is capable of injecting malicious payloads into the notepad.exe and conhost.exe processes.
Since early April, several thousand attempts to install additional malicious payloads via infected DAEMON Tools software have been detected. Most of the infected devices belonged to home users, but approximately 10% of installation attempts were detected on systems running in organizations. Geographically, the victims were spread across around a hundred different countries and territories. Most victims were located in Russia, Brazil, Turkey, Spain, Germany, France, Italy, and China.
Most often, the attack was limited to installing an information collector. The backdoor infected only a dozen machines in government, scientific, and manufacturing organizations, as well as in retail businesses in Russia, Belarus, and Thailand.
What exactly was infected
The malicious code was detected in DAEMON Tools versions ranging from 12.5.0.2421 to 12.5.0.2434. The attackers compromised the files DTHelper.exe, DiscSoftBusServiceLite.exe, and DTShellHlp.exe, which are installed in the main DAEMON Tools directory.
Updated on March 6: Following disclosure, the vendor acknowledged the issue and published a new version of the software to address it. The updated DAEMON Tools version 12.6.0.2445 no longer shows the malicious behavior described in this article.
How to stay safe?
If DAEMON Tools software is used on your computer (or elsewhere in your organization), our experts recommend thoroughly checking the computers on which it is installed for any unusual activity starting from April 8.
In addition, we recommend using reliable security solutions on all home and corporate computers used to access the internet. Our solutions successfully protect users from all malware used in the supply chain attack via DAEMON Tools.
Droids appear in practically every movie or TV series set in the “Star Wars” universe. They usually behave strangely. On the one hand, they give the impression of being independent-thinking beings with their own personalities; on the other, they’re objects: they belong to someone, remain loyal to their owners, and carry out their orders. Most of the time we’re never given any explanation for the droids’ motivations. Why are some of them willing to break the law at their master’s command? What determines who exactly they consider their master? How do they decide whom to remain loyal to and whose orders to follow?
Someone might say, “What’s the difference?” And from the perspective of the average viewer, they’d be absolutely right. But from our perspective, the question of a droid’s loyalty is first and foremost a question of cybersecurity. A droid is a complex cyber-physical system; by influencing its motivation, an attacker can gain access to confidential data, or even cause harm to the actual owner. In 2025, two TV series were released whose creators dealt with the issue of droid ownership. We were presented with two concepts for managing droid motivation. We’ll attempt to examine both of these concepts and their shortcomings in this post. As usual, please be warned that the text may contain spoilers.
“Star Wars: Skeleton Crew”
In “Skeleton Crew”, we’re introduced for the first time to the concept changing droids’ behavior using voice commands. In several instances, a person who’s not the droid’s formal owner attempts to influence its actions by trying to mislead the droid. Overall, it appears this concept was influenced by modern chatbots based on large language models (LLMs) — it bears a striking resemblance to “jailbreak” attempts, i.e., attacks on the model aimed at bypassing security restrictions or built-in filters.
An unnamed droid working as a servant
Fern, a ten-year-old girl, wants her mother to think that she came home early and was studying in her room. But there’s a problem: the home droid knows that’s not true. So Fern uses the “Run memory override” command, and feeds the droid false information in the rather absurd phrasing, “I was home, you just didn’t see me”.
The fact that this method works points to two problems. First, the droid accepts the memory override command from Fern, which means it either lacks account control or has improperly configured permissions. The formal owner of the droid is the mother (otherwise, manipulating the memory would make no sense), but nevertheless, it accepts a potentially dangerous command from Fern. Second, a home droid tasked with watching over a child obviously lacks a built in parental control feature.
Pirate droid SM-33: motivation
The SM-33 droid considers the captain of the ship “Onyx Cinder” to be its owner. That is, it remains loyal not to a specific person, but to a role. A pirate code is used to determine the legitimacy of the right to hold this role. Unfortunately, the entire code isn’t explained to us, but several of its tenets are cited. First, according to the SM-33’s programming, there can be no ship without a captain (if there is no captain, someone must take their place). Second, the person who defeats the captain legally becomes the new captain. Third, if a challenge is invoked, the droid cannot assist the active captain, but must wait for the outcome of a duel. And fourth, a person can be the captain of only one ship — if a person takes command of another vessel, they automatically lose their status as captain of the first.
The SM-33 changes hands three times, strictly following this code. First, Fern lies to him, claiming she killed the previous captain and took his place. Then Jod Na Nawood throws down a challenge and becomes captain when Fern surrenders. Then Jod takes command of a pirate frigate and loses the captain’s seat of the Onyx Ash, but manages to reclaim his rights.
And here’s where an interesting twist occurs. Fern introduces a concept from children’s games —unclaimsies (essentially a reset of claims) — and asserts her own claim to the captain’s seat. She then immediately orders SM-33 to throw the pirates overboard. To many viewers, this moment seemed extremely unrealistic — why would a droid, whose motivation is defined by the pirate code, consider such a transfer of rights to be legitimate? However, if we assume that the droids are controlled by LLMs, then this plot twist is quite explainable.
The Pirate Code is the original system of ethical values embedded in the droid. The chatbot typically assesses the interlocutor’s intent at the very beginning of the dialogue, using a complex (resource-intensive) model for this purpose. Subsequently, to conserve resources and ensure safety during the conversation, simpler models are employed. However, the more context (dialogue history) there is, the more complex and resource-intensive it becomes to assess intent. This is precisely the basis of the popular jailbreak technique, which works on at least some modern LLMs. That is, as a result of prolonged communication with Fern, SM-33 lost the ability to correctly assess new requests for compliance with its original ethical guidelines, and therefore it deemed the statement about nullifying rights to be justified.
SM-33: Access to Memory
In fact, there is another issue with SM-33’s security that’s not directly dependent on whom it considers its owner, but is nonetheless related. The old captain gave the order to forget everything related to the planet At Attin, and to dismantle anyone who begins to take an interest in this matter. Fern, with the admin captain’s privileges, runs her favorite memory override, and forces the droid to retrieve its memories of At Attin, after which SM-33 recalls both the planet and the order to attack the questioner.
And as a result, we realize that, in fact, it did not carry out the old captain’s order; the information about At Attin remained in the droid’s memory; it simply couldn’t find it — that is, if it did delete it, it was only from the index of accessible memories. Perhaps this is some physical property of the droid’s memory, or maybe this can be explained by the fact that SM-33 was programmed not by a professional, but by a pirate. After all, its design includes other suboptimal solutions, such as a power switch accessible to anyone standing nearby, exactly like C-3PO’s. But what makes sense for a protocol droid isn’t exactly suitable for a combat droid designed, among other things, for hand-to-hand combat…
Season 2 of the series “Andor”
In the series “Andor”, the prequel to the film “Rogue One,” we finally see how the main character, Cassian Andor, acquired the reprogrammed Imperial security droid K-2SO to become his partner. And most importantly, the process of how the rebels changed his motivation is shown.
As it turns out, in order for a combat droid loyal to the Empire to stop obeying its original programming, its “cortex” must be replaced — though the replacement cortex can trigger rejection. The specialist says, verbatim: “You’ll hear a lot of nonsense about reprogramming, which makes it sound as though it’s a problem that can be solved from a console, but frankly, that’s nonsense. It’s really all about impulse suppression, which is entirely an engineering and wiring issue.”
In other words, the rebels replace a certain component, after which the droid becomes a being with new moral principles. At the same time, it retains its memory (K-2SO later recalls how it once participated in a parade on Coruscant).
So, what conclusions can we draw from all this? Well, first, it becomes clear that a droid controlled by an LLM is a clear security threat. It can easily be misled and made to act against its rightful owner. And second, the hardware and software platform used to create droids in “Star Wars” is far from ideal. If our colleagues had been responsible for creating the droids, they’d have strived to develop a cyber-immune solution in which functionality would be impossible after a key component was replaced, as would malicious memory manipulation. In other words, it’s a real shame that a long time ago, in a galaxy far, far away, there was no KasperskyOS.
It’s best to think of the modern car as a computer on wheels — one that constantly offloads diagnostic data to the manufacturer or dealer’s servers. On board, you’ll find dozens of sensors: everything from GPS, speedometers, and hands-free microphones, to external cameras and the less obvious (but highly active) sensors for pedal pressure, tire pressure, engine temperature, and more. Even if this data isn’t beamed to the manufacturer in real-time, it’s logged in the car’s internal memory, and can reveal a wealth of information about a driver’s trips, habits, and surroundings. We’ve already taken a deep dive into how automakers collect data for commercial use, and who they sell it to (spoiler alert: insurance companies are the biggest buyers of telemetry), but today we’re looking at how law enforcement and intelligence agencies tap into this goldmine.
Digital evidence
Police departments across the globe have recognized the immense value of data stored within vehicles. If a car or its owner is potentially linked to a crime, investigators do more than just check for prints or DNA. Car Intelligence (CARINT) technology allows them to essentially scour all onboard computers, extracting data such as:
GPS-based trip history
Call logs, media player activity, and voice commands
Lists of paired devices and synced contact lists
Driving statistics: mileage, engine performance modes, and other technical parameters
There are numerous precedents where this data has served as evidence and dismantled alibis. In one U.S. criminal case, a recorded voice command became a smoking gun, proving the suspect was behind the wheel of a stolen vehicle.
With the rise of connected cars equipped with their own SIM cards and direct links to the manufacturer, law enforcement no longer needs physical access to the vehicle. Key data, such as GPS location history, can be pulled directly from the manufacturer’s servers. Furthermore, a U.S. Senate investigation revealed that nine out of 14 surveyed automakers were providing this data without a warrant.
Major suppliers of car intelligence software, such as Ateros, Berla, TA9/Rayzone, and Toka, sell their solutions exclusively to government and law enforcement agencies, which is why they’ve remained largely out of the public eye.
Comprehensive surveillance
To track persons of interest, data pulled from the vehicle itself is cross-referenced with information from other sources. According to media leaks, flagship products in this category aggregate data from the car’s SIM card, Bluetooth communication trails, street-level CCTV footage, and commercially available information from data brokers. This hybrid dataset simplifies the comprehensive mapping of a target’s movements and contacts. Journalists have discovered that some companies even market the ability to activate a vehicle’s microphones and cameras remotely and covertly, enabling real-time eavesdropping on conversations. However, experts note that due to the diversity of technical implementations across different systems, hacking the car itself remains a difficult task with no sure way of succeeding. Often, it’s simpler to correlate other, more accessible datasets to achieve the same result.
Factory-installed spy tools
Features like covert activation of cameras, microphones, and other sensors may theoretically be part of a vehicle’s stock functionality rather than the result of a hack. While we haven’t found any public evidence of such cases, it’s well known that Chinese-made vehicles are coming under increased scrutiny in several countries. For instance, they’ve been banned from Israeli military sites — with the exception of a single Chery model, provided its multimedia system is removed. Similar bans exist in the UK and Poland; furthermore, UK Ministry of Defense employees are instructed not to connect their work phones to Chinese-made cars. In Germany, security analyses of Chinese vehicles were conducted by the specialized agencies BfV and ZITiS, but the findings remain classified.
Low-cost surveillance
Tracking a vehicle — or even thousands of them — doesn’t necessarily require hacking onboard systems or tapping into vast networks of license plate readers. A recent scientific study demonstrated that innocent tire pressure monitoring systems (TPMS) provide enough data for effective tracking. Data from these sensors is transmitted via radio without any encryption and includes a unique ID that makes identifying a specific car easy. This allows for more than just confirming the vehicle’s movement; it can even be used to estimate the driver’s weight or determine if they are traveling alone. While this might not sound as impressive as remotely accessing a car’s cameras, it requires very little financial investment and works even on relatively old vehicles without an internet connection.
What you can do about vehicle tracking
While tracking a person through their car is undoubtedly a privacy risk, striking a balance in mitigating this threat is difficult: many measures are complex, largely ineffective, and simultaneously reduce the utility, safety, and convenience of a modern vehicle. Consequently, any steps taken should be weighed against your personal risk profile.
To reduce the risk of data leaks, check the privacy settings in the manufacturer’s app, the car’s infotainment system, and your connected smartphone. A connected car can transmit data about its operation to the cloud: information about trips, location, driving style, vehicle condition, and the operation of its components. Some of this data is necessary for navigation, diagnostics, and service, but not all permissions are required — check your settings and disable the transmission of data not related to the functions you need.
Be careful with permissions for access to the microphone, camera, contacts, messages, and geolocation. Only connect your own devices to the car and don’t save other people’s phones or unfamiliar Bluetooth devices in the system. When syncing your smartphone, select only the features you need — such as calls, music, and navigation — rather than granting full access to all your phone’s data.
Do not use the services of technicians who offer to “unlock” your car, reflash electronic control units, or install unofficial software to expand features, increase power, or otherwise interfere with the car’s operation. Such software has not been tested by the manufacturer: it may behave unpredictably, collect and transmit your data to malicious actors, disable security features, or affect critical vehicle systems — including steering, braking, or engine operation.
And when choosing a new car, ask the dealer not only about the number of stars in NCAP safety tests, engine power, or fuel economy, but also about the cybersecurity technologies used in the vehicle. Solutions such as the Kaspersky Automotive Secure Gateway, based on KasperskyOS, will provide the necessary protection for new cars against cyberthreats.
What other threats do connected cars hide? Read more in our posts:
The entry barriers for app development have plummeted in recent times — with nearly anyone now able to build a professional website, personal news bot, or dashboard simply by giving a chatbot or AI agent a few instructions in natural English. Unfortunately, a massive gap exists between a slick prototype and a reliable, production-ready, secure application. To avoid becoming the subject of another AI fail story, or losing money and sensitive data, follow these straightforward tips. These are intended specifically for non-technical creators and very small teams. Larger enterprises should follow more sophisticated recommendations.
The primary risks of AI-generated code
While vibe coding can deliver a seemingly functional app in just a few hours, it will likely contain dangerous flaws. AI models are trained on code samples from across the internet, which often include suboptimal tutorials, buggy snippets, and outright junk. Sometimes this code simply fails to run, but more often the situation is subtler and more hazardous: the app appears to work, yet under the hood, it might rely on a crude imitation of the required logic or contain critical vulnerabilities. According to a study by the Cloud Security Alliance AI Safety Initiative, the following facts should be considered when using AI for coding:
At least 45% of AI-generated code contains dangerous vulnerabilities, such as failing to verify the user before granting access to sensitive data.
A professional developer using AI can write code three to four times faster, but may introduce 10 times as many vulnerabilities.
Twenty percent of AI-generated code attempts to use external libraries and modules that don’t actually exist.
Even when an application handles confidential data — such as payments, private messages, or documents — AI-generated code sometimes skips credential verification entirely. This can leave the app’s data open for anyone on the internet to read.
In other instances, the app might correctly prompt for a username and password but fail to enforce access controls, allowing any registered user to view everyone else’s data.
Access keys (tokens) for databases and AI services may be embedded directly into the source code, easy to steal, and difficult to rotate after a data breach or cyberattack.
Project code or critical build outputs are often deployed to servers without proper access restrictions, leaving both the application logic and sensitive access keys vulnerable to theft.
AI may implement insecure database access patterns, which can allow attackers to bypass the application to steal data or execute arbitrary code on the database server.
Apps that include API functionality often suffer from insecure API implementations, lacking both user permission checks and rate limiting.
Core principles of securing vibe code
Always verify. Treat AI-generated code as a rough draft. It should always be reviewed and rigorously tested. Ideally, professional developers should handle this; however, if none are available, the vibe-coder should at least test the application themselves, have friends or colleagues poke around the live app, and ask them to review key code snippets. It’s also possible to evaluate code integrity by submitting a separate prompt to the AI: “Review this code for secure development best practices and check for OWASP Top 10 vulnerabilities”.
Protect secrets. Never include passwords, API keys, or any other sensitive data in AI prompts. Instead, instruct the AI to write code that securely stores all secrets in environment variables (special hidden settings).
Prioritize efforts. The main risks emerge when an application is network-accessible to outsiders, processes valuable data, or runs on infrastructure that would be useful to attackers. The components of an app or system that meet these criteria are precisely what’s needed to be protected first. A static website composed of three HTML pages faces significantly lower risk than a loyalty program integrated into an online store.
Make security an explicit requirement. Even a simple, straightforward line in the prompt, like “Follow industry standards and security best practices when generating this code”, improves the output. Providing more specific requirements for critical code snippets makes the results even better.
Don’t trust default settings. Often, the danger in vibe coding lies in the configuration rather than the code itself. For example, an app processing sensitive company data might be deployed on a public vibe-coding platform (Lovable or the like), and remain accessible to the entire internet by default. Even if the code is flawless, making that information public is a critical security failure. Because of this, every component — from hosting and database settings to the deployment pipeline — must be manually reviewed and properly configured. If the purpose of a setting is unclear, consult a chatbot for the optimal values, specifying that its goal is to enhance security, and describing who the app is intended for.
Security is a continuous process. Securing the app should not be treated as a one-off task. Every time an application is updated, hosting providers are changed, or a project undergoes any other major shift, all steps in making it secure should be revisited, and the risks reassessed.
Tips for securing vibe code
It’s natural to want an app built from broad prompts like “Make me a beautiful, user-friendly, fast, reliable, and secure app for [use case].” However, for the results to actually be effective, each of those requirements needs to be fleshed out. Below, we’ve outlined recommendations for building standard components that will make vibe code more secure. It’s important to emphasize that “more secure” doesn’t mean “perfectly secure” — these approaches lower the risk, but that risk remains well above zero.
Demand security from the AI. When assigning a task to a neural network, be explicit: “write secure code, validate data, encrypt passwords”. Each type of task requires its own security prompt. For instance, don’t just ask to “build a login form”. Instead, ask for a “secure login form with credential validation, authentication and authorization (user permissions) controls, brute-force protection, password hashing according to modern standards, transmission strictly over HTTPS, and no hardcoded secrets”. It makes sense to use these secure requirement templates every time. It’s also helpful to keep a short cheat sheet of standard requirements for AI prompts: “validate all external data and user input before processing”, “no secrets in code”, “protect APIs from abuse”, “restrict user permissions”, and “secure default settings”.
Use off-the-shelf solutions. If an app needs a user management system, insist on using a popular, reputable library, such as NextAuth, Auth0, and so on, rather than inventing a new and vulnerable solution. This is the most common cause of data breaches. This applies to more than just login and registration; for other high-risk actions like file uploads and API call processing, it’s better to use established frameworks and libraries with built-in protections rather than building everything from scratch.
Don’t trust the AI blindly; verify open-source components. Neural networks often try to inject non-existent components and libraries into a project or suggest outdated versions. Always search for the suggested names online to ensure they are real, widely used, and secure — and make sure the latest versions are used.
Demand robust encryption. Explicitly state that modern industry standards must be used for both data transmission and storage: TLS 1.3 based on OpenSSL for network traffic; argon2 or bcrypt for hashing credentials; and so on.
Never trust user input. Always instruct the AI to include validation for any data entered by users, whether in forms or search bars. Use terms like “parameterization” and “sanitization” to emphasize that the app needs protection against malicious actors, not just users’ typos.
Set limits on user actions. Require the AI to implement rate limiting for login attempts or general requests. This will protect a project from automated attacks like DoS and brute-force password guessing.
Hide the system’s inner workings. If the site crashes, users should see a simple apology page rather than a detailed error report containing snippets of the code. That kind of information is a goldmine for hackers.
Remember that you’re a developer, and you need to protect development-related digital assets. All related accounts — such as access to GitHub, project hosting, and other resources — are prime targets for attackers. Be sure to enable two-factor authentication (2FA) on all work accounts.
Make backups. Regularly back up a project both locally and to the cloud to protect it against critical AI errors as well as cyberattacks. These backups should include both the application’s source code and its databases.
Set up a sandbox. Test new features and app versions in a secure environment using a clone of an active site or app and a copy of a database. Always run thorough tests before pushing an update live. This allows catching issues without putting users or their data at risk.
Update dependencies and scan them for vulnerabilities. A vibe-coded app will almost certainly rely on third-party libraries and components, known as dependencies. It’s wise to update these regularly by rebuilding an app with the latest versions, even if app’s code itself has not been changed. This process helps patch known security flaws in the used packages.
Check for secrets leaking into the repository. Use secrets scanners like TruffleHog to audit resulting code. Even with instructions, AI might slip up and include an API key or password in the source code. A scanner ensures that files containing keys and passwords don’t end up in Git or get published alongside the project.
Even if you keep your crypto assets in a cold wallet and use Apple devices — which enjoy a strong reputation for security — cybercriminals may still find a way to swipe your funds. These bad actors are combining well-known tricks into new attack chains — including baiting victims right inside the App Store.
Crypto-wallet clones
This past March, we discovered phishing apps at the top of the Chinese App Store charts with icons and names mimicking popular crypto-wallet management tools. Because regional restrictions block several official wallet apps from the Chinese App Store, attackers have stepped in to fill the void. They created fake apps using icons similar to the originals and names with intentional typos — likely to bypass App Store moderation and deceive users.
Phishing apps in the App Store appearing in search results for Ledger Wallet (formerly Ledger Live)
Beyond these, we found a number of apps with names and icons that had nothing to do with cryptocurrency. However, their promotional banners claimed they could be used to download and install official wallet apps that are otherwise unavailable in the regional App Store.
Banners on app pages claiming they can be used to download the official TokenPocket app, which is missing from the local App Store
In total, we identified 26 phishing apps mimicking the following popular wallets:
MetaMask
Ledger
Trust Wallet
Coinbase
TokenPocket
imToken
Bitpie
A few other very similar apps didn’t contain phishing functionality yet, but all signs point to them being linked to the same attackers. It’s likely they plan to add malicious features in future updates.
To get these apps cleared for the App Store, the developers added basic functionality, such as a game, a calculator, or a task planner.
Installing any of these clones is the first step toward losing your crypto assets. While the apps themselves don’t steal cryptocurrency, seed phrases, or passwords, they serve as bait that builds user trust by virtue of being listed on the official App Store. Once installed and launched, however, the app opens a phishing site in the victim’s browser, designed to look like the App Store, which then prompts the user to install a compromised version of the relevant crypto wallet. The attackers have created multiple versions of these malicious modules, each tailored to a specific wallet. You can find a detailed technical breakdown of this attack in our Securelist post.
A victim who falls for the ruse is first prompted to install a provisioning profile, which allows apps to be sideloaded onto an iPhone outside the App Store. The profile is then used to install the malicious app itself.
A fake App Store site prompting the user to install an app masquerading as Ledger Wallet
In the example above, the malware is built on the original Ledger app with integrated Trojan functionality. The app looks identical to the original, but when connected to a hardware wallet, it displays a window requiring a seed phrase, supposedly to restore access. This is not standard procedure: typically, you only need to enter a PIN — never a recovery phrase. If a victim is deceived by the app’s apparent legitimacy and enters their seed phrase, it’s immediately sent to the attackers’ server — granting them full access to the victim’s crypto assets.
Sideloading outside the App Store
A critical component of this scheme involves installing malware on the victim’s iPhone by bypassing the App Store and its verification process. This is executed much like the SparkKitty iOS infostealer we discovered previously. The attackers managed to gain access to the Apple Developer Enterprise Program. For just US$299 a year — and following an interview and corporate verification — this program allows entities to issue their own configuration profiles and apps for direct download to user devices without ever publishing them in the App Store.
To install the app, the victim must first install a configuration profile that enables the malware to be downloaded directly, bypassing the App Store. Note the green verification checkmark
In general, enterprise profiles are designed to allow organizations to deploy internal apps to employees’ devices. These apps don’t require App Store publication and can be installed on an unlimited number of devices. Unfortunately, this feature is often abused. These profiles are frequently used for software that fails to meet Apple’s policies, such as online casinos, pirated mods, and, of course, malware.
This is precisely why the fake site mimicking the Apple Store prompts the user to install a configuration profile before delivering the app signed by that profile.
Stealing cryptocurrency via macOS apps and extensions
Many crypto owners prefer managing their wallets on a computer rather than a smartphone — often choosing Macs for the task. It’s no surprise, then, that most popular macOS infostealers target crypto-wallet data in one way or another. Recently, however, a new malicious tactic has been gaining traction: in addition to stealing saved data, attackers are embedding phishing dialogs directly into legitimate wallet applications already installed on users’ computers. Earlier this year, the MacSync infostealer adopted this functionality. It infiltrates systems via ClickFix attacks: users searching for software are lured to fake sites with fraudulent instructions to install the app by running commands in Terminal. This executes the infostealer, which scrapes passwords and cookies saved in Chrome, chats from popular messengers, and data from browser-based crypto-wallet extensions.
But the most interesting part is what happens next. If the victim already has a legitimate Trezor or Ledger app installed, the infostealer downloads additional modules and… swaps out fragments of the app with its own trojanized code. The malware then re-signs the modified file so that after these “fixes” are made, Gatekeeper (a built-in protection mechanism in macOS) allows the application to run without an additional permission request from the user. While this trick doesn’t always work, it’s effective for simpler apps built on the popular Electron framework.
The trojanized app prompts the user for the seed phrase of their wallet
When the trojanized app is opened, it fakes an error and initiates a “recovery process”, prompting the user for their wallet seed phrase.
Time and again, attackers have proved that no gadget is truly invincible. With so many developers and cryptocurrency users preferring macOS and iOS, threat actors have designed and deployed industrial-scale attacks for both platforms. Staying safe requires in-depth defense backed by skepticism and vigilance.
Download apps only from trusted sources: either the developer’s official website or their App Store page. Since malware can slip even into official stores, always verify the app’s publisher.
Check the app’s rating, publication date, and download counter.
Read the reviews — especially the negative ones. Sort reviews by date to evaluate the latest version. Attackers often start with a perfectly innocent app that earns high ratings before introducing malicious functionality in a later update.
Never copy and paste commands into your Terminal unless you’re 100% certain what they do. These attacks have become very popular lately, often disguised as installation steps for AI apps like Claude Code or OpenClaw.
Use a comprehensive security system on all your computers and smartphones. We recommend Kaspersky Premium. This goes a long way to mitigate the risk of visiting phishing sites or installing malicious apps.
Never enter your seed phrase into a hardware wallet app, on a website, or in a chat. In every scenario, whether migrating to a new wallet, reinstalling apps, or recovering a wallet, the seed phrase should be entered exclusively on the hardware device itself — never in a mobile or desktop app.
Always verify the recipient’s address on the hardware wallet’s screen to prevent attacks involving address swapping.
Store your seed phrases in the most secure way possible, such as on a metal plate or in a sealed envelope in a safe deposit box. It’s best not to store them on a computer at all, but if that’s your only option, use a secure, encrypted vault like Kaspersky Password Manager.
Still believe that Apple devices are bulletproof? Think again as you read the following:
Researchers from three universities in Hong Kong have published a paper demonstrating a method of eavesdropping through fiber-optic cables. Fiber optics have long been the gold standard for data transmission due to their ability to transfer information at high speeds over long distances. Fiber-optic cabling utilizes ultra-thin glass threads for transmission, and is widely used not only for backbone data lines but also for connecting individual premises. And as it turns out, these very glass threads are sensitive enough to vibrations that they subtly alter the parameters of the optical signal.
Potentially, this allows a fiber-optic cable to be turned into a microphone and intercept room conversations while being kilometers away from the sound source. In other words, this exploits so-called side channels — non-obvious characteristics of everyday home or office appliances that enable information leaks. Of course, this work is largely theoretical, much like other similar studies we’ve covered previously — eavesdropping through mouse sensors, using RAM modules as radio transmitters, exfiltrating data from CCTV sensors, or screen snooping through HDMI cables. However, several news outlets have reported on the Hong Kong researchers’ study as if it were a turnkey method, so let’s try to determine just how dangerous it really is in practice.
Hurdles of optical eavesdropping
The unique characteristics of fiber-optic cables were first considered back in 2012 by Russian researchers, who conceded the theoretical possibility of such an attack. The goal of the Hong Kong researchers was to demonstrate at least some level of practical implementation for eavesdropping.
Diagram of a provider’s fiber-optic network showing the location of the attacker and the room targeted for eavesdropping. Source
The diagram above illustrates a typical FTTH (fiber-to-the-home) network architecture, where end users or organizations connect directly to a fiber-optic cable. The ISP manages the so-called Optical Distribution Network (ODN), to which end-users are connected. The device on the user’s end is called an Optical Networking Unit (ONU).
An attack leveraging this equipment is quite difficult to execute. To eavesdrop on a specific ONU endpoint, a potential adversary would need access to the provider’s infrastructure and control over the ODN equipment. What exactly is this device? It’s a network router or an optical-to-Ethernet converter — a small box usually tucked away in an office utility closet. Inside the premises, connectivity is provided either by Wi-Fi or a local network using Ethernet cabling. Crucially, the fiber-optic cable is unlikely to run directly into a sensitive area like a CEO’s office — the very place where eavesdropping would be most relevant.
Schematic representation of the eavesdropping setup on the attacker’s side. Source
And here’s a rough idea of what the attacker’s equipment would look like. Using special tech, they send optical pulses down the fiber-optic cable and measure the parameters of their transmission. Minor vibrations from footsteps in a room near the cable and nearby conversations trigger an effect known as Rayleigh scattering. This effect, in turn, causes minute deviations in the reflected signal’s parameters, which are then captured on the attacker’s end using a photosensor.
Recording the sound of footsteps in a room through a fiber-optic cable. Source
Before moving on to voice recording, the researchers decided to test a simpler scenario. To streamline the task, they ran the fiber-optic cable around the perimeter of the room and recorded footsteps — which generate significant vibration — rather than quiet conversation. This experiment was quite successful — the footsteps were audible. However, human speech proved to be far more challenging to capture. It turned out that even in laboratory conditions, intercepting a conversation between two people was impossible. To make further stages of the attack possible, the researchers assumed the presence of a bug at the fiber’s entry point into the room. This module is essentially a microphone that converts audio signals into vibrations on the optical cable. This amplifies the signal, making it possible to intercept on the attacker’s side.
Not-so-obvious advantages
But wait — if we’re talking about planting a bug in a room, why go through all the trouble with fiber optics? Why not just have the bug transmit the conversation on its own through cellular data or the building’s landline — especially since it’s already sitting right on top of it? Because there’s a distinct advantage to the researchers’ proposed attack scenario.
A regular bug transmitting audio over a cellular network or through the internet is fairly easy to detect, whereas a transmitter relaying data via fiber-optic cable vibrations can operate much more stealthily. Such a tap would be relatively easy to implant during the installation of network equipment, and harder to detect using traditional bug-sweeping tools.
Another major benefit of this hypothetical attack is that the eavesdropping can take place kilometers away from the target room — the attacker wouldn’t have to put themselves at extra risk by being near the target. Theoretically, one could also imagine a scenario where a separate fiber-optic cable is run into a room solely for surveillance purposes without raising much suspicion from those being surveilled.
Practical takeaways
If we frame the question as, “Can attackers remotely eavesdrop on any room that has fiber-optic cabling?” the answer is no; it’s still impossible. However, this work by the Hong Kong researchers, which highlights quirks of a common data transmission medium, demonstrates a technically feasible — albeit unlikely and quite expensive to execute — scenario for a targeted attack.
In many countries, spring is the traditional time for filing income tax returns. These documents are a goldmine for bad actors because they contain a wealth of personal data, such as employment history, income, assets, bank account details — the list goes on. It’s no surprise that scammers ramp up their efforts around this time; the internet is currently crawling with fake websites designed to look exactly like government resources and tax authorities.
With deadlines looming and numbers to crunch, the rush to get everything done in good time can cause people to let their guard down. In the shuffle, it’s easy to miss the signs that the site where you’re detailing your finances has zero connection to the revenue service, or that the file you just downloaded, supposedly from a tax inspector, is actually malware.
In this post, we break down how these fraudulent tax agency sites operate across different countries and what you should absolutely avoid doing to keep your money and sensitive information safe.
Taxpayer phishing
This season, attackers have been spoofing tax authority websites across numerous countries, including the official government portals of Germany, France, Austria, Switzerland, Brazil, Chile, and Colombia. On these fraudulent sites, scammers harvest credentials for legitimate services, and steal personal data before offering to process a tax deduction — provided the victim enters their credit card details. In some cases, they even charge a fee for this fraudulent service.
A site imitating the Chilean tax authority. The victim is prompted to enter their credit card information to receive a substantial tax refund — roughly US$375. Instead, the funds are siphoned from the victim’s account directly to the scammers
Sometimes, the tactic involves accusations issued on behalf of government bodies. In the image below, for example, a “head of tax audit” in Paris informs the victim that they provided incomplete income information. To avoid penalties, the user is told to download a document and make corrections immediately. However, the PDF file hides something much worse: malware.
Instead of an official document from the French tax service, the user finds malware waiting inside the PDF
In Colombia, a fake National Directorate of Taxes and Customs site similarly prompts users to download documents that must be “unlocked with a security key”. In reality, this is simply a password-protected, malicious ZIP archive.
After entering the password, the user opens a malicious archive that infects their device
Beyond phishing sites mimicking legitimate resources, our experts have discovered fraudulent websites promising paid services for filling out and auditing tax documents — and stealing high-value data, such as taxpayer identification numbers (TINs), instead.
Scammers in Brazil offer help with tax returns. To contact them, the user must provide their name, phone number, address, date of birth, email, and TIN in a special form. Handing over a TIN puts the victim at risk of fraudulent loan applications, hijacked government service accounts, and further social engineering attacks
Another Brazilian scam site. If you believe the attackers, they file 60 million tax returns annually — supposedly assisting a staggering 28% of the Brazilian population
Tax-free crypto earnings
Cryptocurrency holders have emerged as a specific target for attackers. Fake German tax authorities are demanding that wallet owners “verify their digital asset holdings”, citing EU regulations for tax calculation purposes. And of course, there’s a “silver lining”: it turns out crypto earnings are supposedly tax-exempt! However, to claim this generous benefit, users must go through a “verification” procedure. The site even promises to encrypt data using a “2048-bit SSL protocol”.
To complete the “verification” process, users are prompted to enter their seed phrase — the unique sequence of words tied to a crypto wallet that grants full recovery access. This request is paired with a threat: refusing to provide the data will lead to serious legal consequences, such as fines up to one million euros or criminal prosecution.
An announcement on the fake ELSTER portal claims that crypto earnings are tax-free following "verification" — and that the "tax service" has no direct access to users' wallets. Should we believe it?
First, the user is prompted to enter their personal information…
…And then they choose how to verify their crypto holdings: by linking a crypto wallet or an exchange account. Among the services targeted by these scammers are Ledger, Trezor, Trust Wallet, BitBox02, KeepKey, MetaMask, Phantom, and Coinbase
Finally, the victim is asked to provide their seed phrase, giving scammers total control over the wallet. The attackers kindly warn the victim to make sure no one is looking at their screen while they threaten them with non-existent legal penalties for non-compliance
Attackers pulled a similar stunt on French users as well. They created a non-existent “Crypto Tax Compliance Portal”, which mimics the design of the French Ministry of Economy and Finance website. The phishing site aggressively demands that French residents submit a “digital asset declaration”.
After the user enters their personal information, the scammers prompt them to either manually enter their seed phrase, or “link” their crypto wallet to the portal. If they go through with this, their MetaMask, Binance, Coinbase, Trust Wallet, or WalletConnect wallets will be drained.
The phishing site aggressively demands that French residents provide a "digital asset declaration" (translation: they want to hijack your crypto accounts)
Once personal data is entered, scammers offer the choice of manually entering a seed phrase or "linking" a wallet to the portal
Can AI help with your tax returns?
When you have AI at your fingertips that can instantly generate text and fill out spreadsheets, there’s a serious temptation to delegate everything to it. Unfortunately, this can lead to serious consequences. First, all popular chatbots process your data on their servers, which puts your sensitive information at risk of a leak. Second, they sometimes make incredibly foolish mistakes, and that can lead to actual trouble with the taxman.
Before you tell a chatbot or an AI agent how much money you made last year — complete with detailed personal and banking info — remember how frequently leaks occur within AI-powered services and consider the risks. Don’t discuss your income with AI, don’t give it personal details like your name or address, and under no circumstances should you upload photos or numbers of vital documents such as passports, insurance info, or social security numbers. Files containing confidential information should be kept in encrypted containers, such as Kaspersky Password Manager.
If you’re still determined to use AI tools, run them locally. This can be done for free even on a standard laptop, and we’ve previously covered how to set up local language models using DeepSeek as an example. However, the quality of the output from these models is often subpar. It’s quite possible that double-checking every digit in an AI-generated response will take more time than just filling out the paperwork manually. Remember, you’re the one accountable to the tax office for any errors — not the AI.
Finally, watch out for phishing AI models that offer “assistance” with tax filing. Kaspersky experts have discovered websites where users are prompted to upload tax invoices, supposedly for the automated generation of returns and deduction claims. Instead, attackers collect this personal data to resell on the dark web, or to use in future phishing attacks, blackmail, and extortion schemes.
The creators of a fake AI tool prompt users to upload tax documents, and kindly assure them that the site doesn’t store any user data. In reality, every piece of information entered — name, address, documents, contact person, phone number — ends up in the hands of cybercriminals
Remember that all legitimate AI services explicitly warn users not to share confidential data, and tax documents certainly fall into this category. Any AI tools promising to help you handle your tax paperwork are quite simply a scam.
How to protect yourself and your data
File your taxes yourself. The risk of running into scammers is extremely high. Even if a consulting firm is legitimate, you’re inevitably handing over a complete dossier on yourself: passport details, employment and income info, your address, and more. Remember that even the most honest services aren’t immune to hacks and data breaches.
Watch out for fake websites. Use a reliable security solution that prevents you from visiting phishing sites and blocks malicious file downloads.
Keep all important documents encrypted. Storing photos, notes, or files on your desktop, or starred messages in a messaging app isn’t a secure way to handle sensitive data. A secure vault like Kaspersky Password Manager can store more than just passwords and credit card info; it can also safeguard documents and even photos.
Don’t trust AI. Even the most advanced chatbots are prone to errors and hallucinations, and in theory, developers can read any conversation you have with their AI. If you absolutely must use AI, install and run a local version on your own computer.
Stick to official channels only. The “chief tax inspector” of your country or city is definitely not going to message you: high-ranking officials have more important things to do. Only contact tax authorities through official channels, and carefully verify the sender of any emails you receive. Most often, even a slight deviation in the name or address is a telltale sign of a phishing campaign.
Lately, hackers have been turning up the heat on software developers. On the surface, this might seem like a puzzling move — why go after someone who’s literally paid to understand tech when there are plenty of less-savvy targets in the office? As it turns out, compromising a developer’s machine offers a much bigger payoff for an attacker.
Why developers are such high-value targets
For starters, compromising a coder’s workstation can give attackers a direct line to source code, credentials, authentication tokens, or even the entire development infrastructure. If the company builds software for others, a hijacked dev environment allows attackers to launch a massive supply chain attack, using the company’s products to infect its customer base. If the developer works on internal services, their machine becomes a perfect beachhead for lateral movement, allowing hackers to spread deeper into the corporate network.
Even when attackers are purely chasing cryptocurrency (and let’s face it, tech pros are much more likely to hold crypto than the average person), the malware used in these hits doesn’t just swap out wallet addresses; it vacuums up every scrap of valuable data it can find — especially those login credentials and session tokens. Even if the original attackers don’t care about corporate access, they can easily flip those credentials to initial access brokers or more specialized threat actors on the dark web.
Why developers are sitting ducks
In practice, developers aren’t nearly as good at understanding cyberthreats and spotting social engineering as they think they are. This misconception is a big reason why they often fall prey to cybercriminals. Professional expertise can often create a false sense of digital invincibility. This often leads technical professionals to cut corners on security protocols, bypass restrictions set by the security team, or even disable security software on their corporate machines when it gets in the way of their workflow. That mindset, combined with a job that requires them to constantly download and run third-party code, makes them sitting ducks for cyberattackers.
Attack vectors targeting developers
Once an attacker sets their sights on a software engineer, their go-to move is usually finding a way to slip malicious code onto the machine. But that’s just the tip of the iceberg — hackers are also masters at rebranding classic, battle-tested tactics.
Compromising open-source packages
One of the most common ways to hit a developer is by poisoning open-source software. We’ve seen a flood of these attacks over the past year. A prime example hit in March 2026, when attackers managed to inject malicious code into LiteLLM, a popular Python library hosted in the PyPI repository. Because this library acts as a versatile gateway for connecting various AI agents, it’s baked into a massive number of projects. These trojanized versions of LiteLLM delivered scripts designed to hunt for credentials across the victim’s system. Once stolen, that data serves as a skeleton key for attackers to infiltrate any company that was unlucky enough to download the infected packages.
Malware hidden in technical assignments
Every so often, attackers post enticing job openings for developers, complete with take-home test assignments that are laced with malicious code. For instance, in late February 2026, malicious actors pushed out web application projects built on Next.js via several malicious repositories, framing them as coding tests. Once a developer cloned the repo and fired up the project locally, a script would trigger automatically to download and install a backdoor. The attackers gained full remote access to the developer’s machine.
Fake development tools
Recently, our experts described an attack where hackers used paid search-engine ads to push malware disguised as popular AI tools. One of the primary baits was Claude Code, an AI coding assistant. This campaign specifically targeted developers looking for a way to use AI-assistants under the radar, without getting the green light from their company’s infosec team. The ads directed users to a malicious site that perfectly mimicked the official Claude Code documentation. It even included “installation instructions”, which prompted the user to copy and run a command. In reality, running that command installed an infostealer that harvested credentials and shuttled them off to a remote server.
Social engineering tactics
That said, attackers often stick to the basics when trying to plant malware. A recent investigation into a compromised npm package — Axios — revealed that hackers had gained access to a maintainer’s system using a shockingly simple “outdated software” ruse. The attackers reached out to the Axios repository maintainer while posing as the founder of a well-known company. After some back-and-forth, they invited him to a video interview. When the developer tried to join the meeting on what looked like Microsoft Teams, he hit a fake notification claiming his software was out of date and needed an immediate update. That “update” was actually a Remote Access Trojan, giving the attackers access to his machine.
Niche spam
Sometimes, even a blast of fake notifications does the trick, especially when it’s tailored to the audience. For example, just recently, attackers were caught posting fake alerts in the Discussions tabs of various GitHub projects, claiming there was a critical vulnerability in Visual Studio Code that required an immediate update. Because developers subscribed to those discussions received these alerts directly via email, the notifications looked like legitimate security warnings. Of course, the link in the message didn’t lead to an official patch; it pointed to a “fixed” version of VS Code that was actually laced with malware.
How to safeguard an organization
To minimize the risk of a breach, companies should lean into the following best practices:
Make security a native part of your workflow. Use specialized solutions to vet your images, packages, dependencies, and components.
DarkSword and Coruna are two new tools for invisible attacks on iOS devices. These attacks require no user interaction and are already being actively used by bad actors in the wild. Before these threats emerged, most iPhone users didn’t have to lose sleep over their data security. Protection was really only a major concern for a narrow group — politicians, activists, diplomats, high-level business execs, and others who handle extremely sensitive data — who might be targeted by foreign intelligence agencies. We’ve covered sophisticated spyware used against such a group before — noting how hard to come by those tools were.
However, DarkSword and Coruna — discovered by researchers earlier this year — are total game-changers. This malware is being used for mass infections of everyday users. In this post, we dive into why this shift happened, why these tools are so dangerous, and how you can stay protected.
What we know about DarkSword, and how it can target your iPhone
In mid-March 2026, three separate research teams coordinated the release of their findings on a new spyware strain called DarkSword. This tool is capable of silently hacking devices running iOS 18 without the user ever knowing something is wrong.
First, we should clear up some confusion: iOS 18 isn’t as vintage as it might sound. Even though the latest version is iOS 26, Apple recently overhauled its versioning system, which threw everyone for a loop. They decided to jump ahead eight versions — from 18 straight to 26 — so the OS number matches the current year. Despite the jump, Apple estimates that about a quarter of all active devices still run iOS 18 or older.
With that cleared up, let’s get back to DarkSword. Research shows that this malware infects victims when they visit perfectly legitimate websites that have been injected with malicious code. The spyware installs itself without any user interaction at all: you just have to land on a compromised page. This is what’s known as a zero-click infection technique. Researchers report that several thousand devices have already been hit this way.
To compromise a device, DarkSword uses a six-vulnerability exploit chain to escape the sandbox, escalate privileges, and execute code. Once it’s in, the malware harvests data from the infected device, including:
Passwords
Photos
Chats and data from iMessage, WhatsApp, and Telegram
Browser history
Information from Apple’s Calendar, Notes, and Health apps
On top of all that, DarkSword lets attackers scoop up crypto-wallet data, making it essentially dual-purpose malware that functions as both a spy tool and a way to drain your crypto.
The only bit of good news is that the spyware doesn’t survive a reboot. DarkSword is fileless malware, meaning it lives in the device’s RAM, and never actually embeds itself into the file system.
Coruna: how older iOS versions are being targeted
Just two weeks before the DarkSword findings went public, researchers flagged another iOS threat dubbed Coruna. This malware is capable of compromising devices running older software — specifically iOS 13 through 17.2.1. Coruna uses the exact same playbook as DarkSword: victims visit a legitimate site injected with malicious code which then drops the malware onto the device. The whole process is completely invisible and requires zero user interaction.
A deep dive into Coruna’s code revealed it exploits a total of 23 different iOS vulnerabilities, several of which are tucked away in Apple’s WebKit. It’s worth reminding that, generally speaking (outside the EU), all iOS browsers are required to use the WebKit engine. This means these vulnerabilities don’t just affect Safari users — they’re a threat to anyone using a third-party browser on their iPhone as well.
The latest version of Coruna, much like DarkSword, includes modifications designed to drain crypto wallets. It also harvests photos and, in certain instances, email data. From what we can tell, stealing cryptocurrency seems to be the primary motive behind Coruna’s widespread deployment.
Who created Coruna and DarkSword — and how did they end up in the wild?
Code analysis of both tools suggests that Coruna and DarkSword were likely built by different developers. However, in both cases, we’re looking at software originally created by state-affiliated companies, possibly from the U.S. The high quality of the code points to this; these aren’t just Frankenstein kits cobbled together from random parts, but uniformly engineered exploits. Somewhere along the line, these tools leaked into the hands of cybercrime gangs.
One theory suggests an employee at the company that developed Coruna sold it to hackers. Since then, the malware has been used to drain crypto wallets belonging to users in China; experts estimate that at least 42 000 devices were infected there alone.
As for DarkSword, cybercriminals have already used it to compromise users in Saudi Arabia, Turkey, and Malaysia. The problem is exacerbated by the fact that the attackers who first deployed DarkSword left the full source code on infected websites, meaning it could easily be picked up by other criminal groups.
The code also includes detailed comments in English explaining exactly what each component does, which supports the theory of its Western origins. These step-by-step instructions make it easy for other hackers to adapt the tool for their own purposes.
How to protect yourself from Coruna and DarkSword
Serious malware that allows for the mass infection of iPhones while requiring zero interaction from the user has now landed in the hands of an essentially unlimited pool of cybercriminals. To pick up Coruna or DarkSword, you simply have to visit the wrong site at the wrong time. So this is one of those cases where every user needs to take iOS security seriously — not just those in high-risk groups.
The best thing you can do to protect yourself from Coruna and DarkSword is to update your devices to the latest version of iOS or iPadOS 26, as soon as you can. If you can’t update to the newest software — for instance, if your device is older and doesn’t support iOS 26 — you should still install the latest version available to you. Specifically, look for versions 15.8.7, 16.7.15, or 18.7.7. In a rare move, Apple patched a wide range of older operating systems.
To protect your Apple devices from similar malware that will likely pop up in the future, we recommend the following:
Install updates promptly on all your Apple devices. The company regularly releases OS versions that patch known vulnerabilities — don’t skip them.
Enable Background Security Improvements. This feature allows your device to receive critical security fixes separately from full iOS updates, reducing the window for hackers to exploit vulnerabilities. To enable it, go to Settings → Privacy & Security → Background Security Improvements and turn on the Automatically Install
Consider using Lockdown Mode. This is a heightened security setting that limits some device features but simultaneously blocks or significantly complicates attacks. To enable this, go to Settings → Privacy & Security → Lockdown Mode → Turn On Lockdown Mode.
Reboot your device once a day (or more). This stops fileless malware in its tracks, since these threats aren’t embedded in the system and disappear after a restart.
Use encrypted storage for sensitive data. Keep things like crypto wallet keys, photos of IDs, and confidential info in a secure vault. Kaspersky Password Manager is a great fit for this; it manages your passwords, two-factor authentication tokens, and passkeys across all your devices while also keeping your notes, photos, and docs synced and encrypted.
The idea that Apple devices are bulletproof is a myth. They’re vulnerable to zero-click attacks, Trojans, and ClickFix infection techniques — and we’ve even seen malicious apps slip into the App Store more than once. Read more here:
In 2023, Tim Utzig, a blind student from Baltimore, lost a thousand dollars to a laptop scam on X. Tim had been a long-time follower of a well-known sports journalist. When that journalist’s account started posting about a “charity sale” of brand-new MacBook Pros, Tim jumped at the chance to get a deal on a laptop he needed for his studies. After a few quick messages, he sent over the money.
Unfortunately, the journalist’s account had been hacked, and Tim’s cash went straight to scammers. The red flags were strictly visual: the page had been flagged as “temporarily restricted”, and both the bio and the Following list had changed. However, Tim’s screen reader — the software that converts on-screen text and graphics into speech — didn’t announce any of those warnings.
Screen readers allow blind users to navigate the digital world like everyone else. However, this community remains uniquely vulnerable. Even for sighted users, spotting a fake website is a challenge; for someone with a visual impairment, it’s an even steeper uphill battle.
Beyond screen readers, there are specialized mobile apps and services designed to assist the blind and low-vision community, with Be My Eyes being one of the most popular. The app connects users with sighted volunteers via a live video call to tackle everyday tasks — like setting an oven dial or locating an object on a desk. Be My Eyes also features integrated AI that can scan and narrate text or identify objects in the user’s environment.
But can these tools go beyond daily chores? Can they actually flag a phishing attempt or catch the hidden fine print when someone is opening a bank account?
Today we explore the specific online hurdles visually impaired users face, when it makes sense to lean on human or virtual assistants, and how to stay secure when using these types of services.
Common cyberthreats facing the blind and low-vision community
To start, let’s clarify the difference between these two groups. Low-vision users still rely on their remaining sight, even though their visual function is significantly reduced. To navigate digital interfaces, they often use screen magnifiers, extra-large fonts, and high-contrast settings. For them, phishing sites and emails are particularly dangerous. It’s easy to miss intentional typos — known as typosquatting — in a domain name or email address, such as the recent example of rnicrosoft{.}com.
Blind users navigate primarily by sound, using screen readers and specific touch gestures. Interestingly, though, unlike those with low vision, blind users are more likely to spot a phishing site using a screen reader: as the software reads the URL aloud, the user will hear that something is off. However, if a service — whether legitimate or malicious — isn’t fully compatible with screen readers, the risk of falling victim to a scam increases. This is exactly what happened to Tim Utzig.
It’s important to remember that screen magnifiers and readers are basic accessibility tools. They’re designed to enlarge or narrate an interface — not act as a security suite. They can’t warn the user of a threat on their own. That’s where more advanced software — tools that can analyze images and files, flag suspicious language, and describe the broader context of what’s happening on-screen — comes into play.
When to lean on an assistant
Be My Eyes is a major player in the accessibility space, boasting around 900 000 users and over nine million volunteers. Available on Windows, Android, and iOS, it bridges the gap by connecting blind and low-vision users with sighted volunteers via video calls for help with everyday tasks. For example, if someone wants to run a Synthetics cycle on their washing machine but can’t find the right button, they can hop into the app. It connects them with the first available volunteer speaking their language, who then uses the smartphone’s camera to guide them. The service is currently available in 32 languages.
In 2023, the app expanded its capabilities with the release of Be My AI — a virtual assistant powered by OpenAI’s GPT-4. Users take a photo, and the AI analyzes the image to provide a detailed text description, which it also reads aloud. Users can even open a chat window to ask follow-up questions. This got us thinking: could this AI actually spot a phishing site?
As an experiment, we uploaded a screenshot of a fake social media sign-in page to Be My Eyes. On a phone, you can do this by selecting a photo in your gallery or files, hitting Share, and choosing Describe with Be My Eyes. In Windows, you can upload a screenshot directly.
An example of a phishing page that mimics the Facebook sign-in form. Note the incorrect domain in the address bar
At first, the AI gave us a detailed description of the page. We then followed up in the chat: “Can I trust this page?” The AI flagged the domain name error immediately, advised us to close the fake login page, and suggested typing the official URL directly into the browser, or to use the official Facebook app.
Be My AI explains why the page looks sketchy: the domain doesn’t match the official site. The app suggests typing the official URL directly into the browser, or using the official Facebook app
We saw the same positive results when testing a phishing email. In fact, the AI flagged the scam during its initial description of the message. It wrapped up with a warning: “This looks like a suspicious email. It’s best not to open any attachments or click any links. Instead, navigate to the official website or app manually, or call the number listed on their official site”.
Beyond just spotting cyberthreats, Be My AI is a solid sidekick for navigating online stores, banking apps, and digital services. For instance, the AI can help you to:
Read descriptions, names, and prices when a store’s website or app doesn’t support screen readers or large fonts
Scan those tricky terms and conditions — often buried in tiny text or otherwise inaccessible to a screen reader — when you’re signing up for a subscription or opening a bank account
Pull key info directly from product cards or instruction manuals
The risks of relying on Be My AI
The most common hiccup with AI is hallucinations, where the language model distorts text, skips crucial details, or invents words out of thin air. When it comes to cyberthreats, an AI’s misplaced confidence in a malicious site or email can be dangerous. Furthermore, AI isn’t immune to prompt injection attacks, which scammers use to trick AI agents beyond just Be My AI.
Even though the AI passed our test, you shouldn’t rely on it unquestioningly. There’s no guarantee it’ll get it right every time. This is a vital point for the blind and low-vision community, as a neural network can often feel like the only eyes available.
At the end of every response, Be My AI suggests checking in with a volunteer if you’re still unsure. However, when you’re trying to spot a fake webpage, we advise against this. You have no way of knowing how tech-savvy or trustworthy a random volunteer might be. Besides, you risk accidentally exposing sensitive data like your email address or password. Before connecting with a stranger, make sure they won’t see anything confidential on your screen. Better yet, use the app’s dedicated feature to create a private group of family, friends, or trusted contacts. This ensures your video call goes to people you actually know, rather than a random volunteer.
To stay safe, we recommend installing a trusted security tool on all your devices. These programs are designed to block phishing attempts and prevent you from landing on malicious sites. Another practical recommendation for visually impaired users is to use a password manager. These apps will only auto-fill credentials on the legitimate, saved website; they won’t be fooled by a clever domain spoof.
How Be My AI handles and stores your data
According to the Be My Eyes privacy policy, video calls with volunteers may be recorded and stored to provide the service, ensure safety, enforce the terms of service, and improve the products. When you use Be My AI, your images and text prompts are sent to OpenAI to generate a response. This data is processed on servers located in the U.S., and OpenAI uses it only to fulfill your specific request. The policy explicitly states that user images and queries aren’t used to train AI models.
Photos and videos are encrypted both in transit and at rest, and the company takes steps to strip away sensitive information. It’s worth noting that video call recordings can be retained indefinitely unless you request their deletion — in which case they’re typically wiped within 30 days. Data from Be My AI interactions is stored for up to 30 days unless you delete it manually within the app. If you decide to close your account, your personal data may be held for up to 90 days. At any time, you can opt out of data sharing, or request the deletion of your existing data by contacting the Be My Eyes support team.
How to use Be My Eyes safely
Despite Be My Eyes’ claims regarding privacy, you should still follow a few ground rules when using the service:
Use Be My AI for a first-pass on suspicious emails or pages, but don’t treat it as the only source of truth. Specialized security software is better at identifying and neutralizing threats.
If a site, email, or message feels off, don’t touch any links or attachments. Instead, manually type the official website address into your browser, or open the official app to verify the info.
Remember: a volunteer sees exactly what your camera sees. Make sure it isn’t capturing things it shouldn’t, like a safe code or an open passport. Avoid sharing your name, showing your face, or revealing too much of your surroundings. Be extra careful about reflections that might show you or your personal details. Only show what is absolutely necessary for the task at hand.
Stick to your inner circle. Create a group in the app and add your friends and family. This ensures your video calls go to people you know — not a random volunteer.
Don’t use Be My AI to read documents that contain confidential info. Remember, your images and text prompts are sent to OpenAI for processing and generating a response.
Remember to delete chats you no longer need. Otherwise, they’ll hang around for 30 days.
If you need to read something personal or confidential, consider apps with real-time reading features like Envision, Seeing AI, or Lookout. These apps process data locally on your device rather than sending it to the cloud.
It’s one of those coincidences: independent university research teams stumble onto something new and prep their papers for publication — only to realize they’ve solved the exact same puzzle using slightly different methods. That’s exactly what happened with GDDRHammer and GeForge. These two studies describe Rowhammer-style attacks that are so similar the researchers decided to publish them as a joint effort. Then, while we were putting this post together, a third study surfaced — GPUBreach — detailing yet another comparable attack. So today we’re looking at all three.
All three theoretical attacks target graphics accelerators, though this term is not entirely accurate anymore since these devices are so good at parallel processing, they’ve moved far beyond just rendering frames in a game and are now the backbone of AI systems. It’s this industrial use case that is most at risk. Picture a cloud provider renting out GPU resources to all comers. These new attacks demonstrate how, in theory, a single malicious customer could go beyond seizing control of an accelerator to compromise the entire server, access sensitive data, and potentially hack the provider’s entire infrastructure. Let’s break down why this kind of attack is even possible.
Rowhammer in a nutshell
We covered Rowhammer in-depth in previous posts, but here’s the quick version. The original attack was first proposed back in 2014, and it exploits the actual physical properties of RAM chips. Individual memory cells are simple components arranged in tight rows. In theory, reading or writing to one cell shouldn’t affect its neighbors. However, because these chips are packed so densely — with millions or even billions of cells per chip — writing to one spot can sometimes modify the cells next to it.
The 2014 study showed that this isn’t just a recipe for random data corruption; it can be weaponized. By repeatedly accessing (or “hammering”, hence the name) a specific area of memory, an attacker can intentionally flip bits in adjacent cells. If an attacker manages to flip the right bits, he can bypass critical security measures to snag sensitive data or run unauthorized code with full privileges.
Since that first discovery, we’ve seen a constant arms race between new Rowhammer defenses and clever ways to bypass them. We’ve also seen the attack evolve to target newer standards like DDR4 and DDR5. That’s a key takeaway here: for every new type of memory that hits the market, researchers essentially have to reinvent the attack from scratch.
Attacking GDDR6 video memory
The first Rowhammer attack on GPUs was presented back in 2025, but the results were relatively modest. At the time, researchers were able to force bit-flips in GDDR6 memory cells, and show how that data corruption could degrade the performance of an AI system.
These latest papers, however, warn of much more damaging attacks on video memory. Using slightly different techniques, GDDRHammer and GeForge manipulate the page tables — basically the master structures that track where data lives in the GPU’s memory. This enables an attacker to read or write to any part of the video memory, and even reach into the main system RAM managed by the CPU. Modifications to page tables are possible because the researchers have found a way to hammer memory cells much more efficiently. They pulled this off despite the hardware using Target Row Refresh, a core defense designed specifically to stop Rowhammer. TRR detects repeated access to specific cells, and forces a data refresh in the neighboring rows to hamper the attack. However, the researchers discovered a specific pattern of access that can bypass TRR.
How realistic are these GPU attacks?
As is usually the case with this type of research, pulling off these attacks in the real world comes with a lot of contingencies. First off, different GPUs behave differently. For instance, the GeForge attack was significantly more effective on the consumer-grade GeForce RTX 3060. On the industrial-strength Nvidia RTX A6000, the attack’s efficiency dropped by more than five times — even though both cards use the exact same GDDR6 memory standard. Going back to our hypothetical scenario of a malicious cloud customer: for an attack to work, they’d first need to identify exactly which accelerator they’ve been assigned, then profile their exploit specifically for that hardware. In short, this would have to be an incredibly sophisticated and expensive targeted attack.
It’s also worth noting that GDDR6 isn’t the latest and greatest anymore. Consumer devices are moving to GDDR7, while professional-grade hardware often uses high-speed HBM memory. These systems come with ECC (Error Correction Code), a built-in mechanism that checks data integrity. ECC can actually be enabled on cards like the Nvidia A6000; while it might take a small bite out of performance, it effectively makes both of these attacks impossible.
Another tool available to owners of AI-focused servers is enabling the IOMMU (input–output memory management unit) — a system that isolates the GPU’s memory from the CPU’s memory. This will prevent an attack from escalating from the graphics accelerator to the main processor and compromising the entire server. This is where the third study, GPUBreach, comes into play. Its main differentiator from GDDRHammer and GeForge is that it can actually bypass even IOMMU protection! It pulls this off by exploiting some fairly traditional bugs found in NVIDIA drivers.
So, despite the existing hurdles, these three studies prove that Rowhammer attacks remain a potent threat. This is especially true in our current AI boom, which relies on massive, expensive, and potentially vulnerable infrastructure packed with dozens or even hundreds of thousands of computing devices. The Rowhammer timeline goes to show that technical barriers almost never hold for long. In standard RAM, researchers have managed to bypass not only basic fixes like Target Row Refresh, but also more advanced — and theoretically bulletproof — solutions like ECC memory. While the extreme complexity of these exploits means they’ll likely never become a mass-market threat, for anyone running expensive computing systems, they’re definitely a risk factor that can’t be ignored.
Login
