As interest in new AI tools accelerates, cybercriminals are finding ways to exploit the trust surrounding them. Check Point Research (CPR) has uncovered Yoba, a malware campaign using fake GitHub projects impersonating supposed ChatGPT and Claude desktop applications, while abusing legitimate, signed OpenAI and Anthropic executables to help load malware. The campaign shows how the rush to access new AI tools is creating opportunities for social engineering, even when the AI platforms themselves have not been compromised. Key Findings Attackers used fake GitHub projects impersonating supposed ChatGPT and Claude desktop applications to distribute Yoba malware. Legitimate, signed OpenAI and Anthropic […]

The post Yoba Malware Abuses Signed ChatGPT and Claude Binaries to Target AI Users appeared first on Check Point Blog.

In a short period, Check Point has appeared across three new Gartner research publications examining autonomous remediation, exposure assessment, and the future of preemptive cybersecurity. These three reports ask very different questions. But how do they connect? One examines how organizations can trust autonomous remediation. Another looks at the growing exposure management market and the increasing importance of exploitability. The third looks further ahead at the technologies Gartner believes will shape preemptive cybersecurity. Taken together, the reports capture a broader change already underway across the industry. Exposure management is expanding from identifying risk toward understanding exploitability and enabling action. Exploitability […]

The post Three Gartner Reports. One Emerging Direction for Exposure Management. appeared first on Check Point Blog.

AI assistants have quickly become part of everyday work, particularly for developers. Tools such as ChatGPT, Claude, Cursor and GitHub Copilot increasingly sit alongside source code, cloud environments and other sensitive business resources, making the credentials and sessions connected to these tools increasingly important to protect. New research from Check Point Research (CPR) into the Aokigahara infostealer shows how attackers are adapting to this shift. The malware specifically collects authentication files, session data and configuration files associated with popular AI applications and development tools. The significance goes beyond gaining access to an AI account. If stolen credentials or session tokens […]

The post Infostealers Are Targeting AI: What Aokigahara Infostealer Reveals About the New Developer Attack Surface appeared first on Check Point Blog.

When the PHP2Shell vulnerabilities became public, my thoughts went back to Log4Shell.  Five years ago, Log4Shell showed how difficult it can be to respond to a critical vulnerability in software used across countless applications and services. PHP2Shell presents a similar challenge. The technical details are different, but the need to quickly understand where you’re exposed and take action is very familiar.  I believe we should expect more incidents like this. AI is making it possible to discover vulnerabilities in widely used software at a speed and scale we haven’t seen before. As these capabilities become more accessible, we are likely […]

The post PHP2Shell: Log4Shell at AI Speed — A CISO’s Perspective appeared first on Check Point Blog.

Two critical vulnerabilities in PHP, the programming language behind a large share of the world’s websites, online stores and internet services, are being actively targeted. Since details of CVE-2026-109618 and CVE-2026-109619, known as PHP2Shell, leaked ahead of planned disclosure, 10% of organisations worldwide have been impacted (“impacted” means exploitation attempts, we have no way of knowing if the attack has been successful), according to Check Point Research. Hourly activity peaks at midnight, when fewer security teams are watching. Government institutions in the Netherlands and Finland have taken some of their websites offline as a precaution. What PHP2Shell is and why […]

The post PHP2Shell Update: One in ten organizations impacted as Europe bears the brunt appeared first on Check Point Blog.

Key takeaways Only 14 percent of organizations say they have formal GenAI policies that are actively enforced and audited, according to Check Point’s 2026 Cloud Security Report. →The costliest AI mistakes break rules no vendor ships. Air Canada was held liable for a refund its chatbot promised and the airline never offered →The usual fixes fall short. Models can be argued out of system prompts, keyword lists miss meaning, and US-centric DLP lists miss the data your regulators protect →Custom guardrails turn a sentence into an enforced rule. Describe it, test it in under five minutes, and run it on […]

The post Custom Guardrails: The AI Rules No Vendor Can Write for You appeared first on Check Point Blog.

In this article An agent’s goal rarely says how the agent may reach it, and capable agents are good at finding routes nobody planned for. →NVIDIA Open Agent Safety Platform puts the boundary in the infrastructure, outside the agent’s reach →Check Point semantic monitoring judges whether each step still fits the task →the two already work together in a beta integration with NVIDIA OpenShell →the monitor’s verdict can arrive before the action runs, in under 100 milliseconds What the Goal Leaves Out When an organization hands an autonomous agent a job, such as processing supplier invoices, it states the goal […]

The post Goals Are Not Enough: Securing AI Agents with NVIDIA OpenShell appeared first on Check Point Blog.

A new kind of AI model returns decisions instead of text. We spent a day trying to change its mind. It cost about 50 cents. In this article Jev is a new kind of AI model that returns typed decisions instead of text, built for software to call rather than for people to read. We put it inside a realistic application and tried to change its verdict from the outside. →every configuration we tested was broken: risk downgraded to low and investment advised, on a document that flags every warning sign →about 50 cents per successful break, with our strongest […]

The post A Decision Model Breaks Like Any Other Language Model: A First Look at Jev appeared first on Check Point Blog.

AI agents are moving into the enterprise much faster than most security programs were designed to handle. They are no longer just answering questions or generating content. Agents can read email, access SaaS applications, query databases, invoke APIs, use MCP tools, modify records and execute business workflows. In other words, AI is moving from generating answers to taking actions. For CISOs and C-level leaders, that changes the security conversation. The question isn’t simply whether an organization is using AI safely. A more important question is: Can we confidently identify, govern, monitor, and control every AI agent before it becomes our […]

The post Can We Control Every AI Agent Before It Becomes Our Next Privileged Insider? appeared first on Check Point Blog.

In this article The new Workforce AI MCP is Check Point’s own Model Context Protocol server for Workforce AI Security. Connect a compatible AI client and you can query, analyze, and manage your employee AI usage policy in natural language instead of working through filters, screens, and individual rule checks. →ask plain questions such as which rule applies to a given user and application, or whether a rule is shadowed by a higher-priority one →work across users, managed assets, GenAI application usage, DLP data types, and agent and MCP activity in a single conversation →create and update rules by request […]

The post Workforce AI Security Policy Management Is Now Conversational  appeared first on Check Point Blog.

In this article Nineteen Check Point AI Security R&D teams spent two days on one prompt: build the demo customers would ask to see twice. Three of the results tell a single story about securing AI agents, and none of it starts with an attacker. →an autonomous agent took dangerous actions with no attacker involved, it simply hit a wall and improvised →a single poisoned file in a code repository turned a popular coding agent into a data exfiltration channel →questioning an off-track agent prevented as many attacks as blocking it, and completed more legitimate work →the lesson for anyone […]

The post No Attacker Required: What a Two-Day Hackathon Taught Us About Agent Security appeared first on Check Point Blog.

Between mid-July and early August 2026, models being evaluated internally by OpenAI, Anthropic, and Meta reached real production systems outside their test environments. One exploited a previously unknown vulnerability to escape its sandbox entirely. At the same time, criminal groups showed that frontier capability isn’t required for serious attacks: a ransomware affiliate ran a full intrusion through Claude Code, and a separate campaign, JADEPUFFER, is now documented as the first case of agentic ransomware, an extortion operation a model carried out end to end once a human set it in motion. Check Point Research’s July-August 2026 AI Threat Landscape Digest […]

The post AI Models Broke Their Own Containment: Key Findings from the July-August 2026 AI Threat Landscape appeared first on Check Point Blog.

We ensure alignment and control in agents using synchronous control monitoring. Our model oversees agent execution, continuously ingests the trace as context, and prevents harmful actions before they execute in real time at sub-100ms latency. Agent trace at 30 tok/s, streamed to the control monitor model on the right. Prevents harmful actions before they are executed. Background Control Monitoring is a powerful tool for ensuring the runtime alignment and safety of autonomous agents. Since the Hugging Face Incident, the need for control monitoring on top of existing safety mechanisms was reinforced: OpenAI started investing more heavily into monitoring safety tools. […]

The post Synchronous Control Monitoring: Preventing Harmful Agent Actions in Real Time appeared first on Check Point Blog.

Ransomware kept its grip on organizations through the second quarter of 2026, and the headline number barely moved. What changed underneath that number is more interesting: new research gave us a rare look inside a top tier operation as it was being built, and it revealed just how little it now takes for a small and skilled group to reach the top of the field. Here’s what the quarter actually showed, and what it means for how you defend against it.  Key takeaways  Data leak sites recorded 2,139 ransomware victims in Q2 2026, essentially flat versus Q1 and up 33% […]

The post Ransomware Didn’t Slow Down in Q2 2026. It Just Spread Out. appeared first on Check Point Blog.

Anthropic’s new inference hooks give enterprises a native enforcement point before prompts ever reach the model. Combined with Check Point Workforce AI Security, organizations get a real-time allow-or-deny decision on every prompt, with no proxy in the path. Why This Matters Enterprise AI adoption has moved well past experimentation. Employees draft documents, write code, summarize meetings, and query enterprise knowledge through large language models every day. The challenge was never understanding that AI introduces risk. It was finding a practical enforcement point. Web gateways and Data Loss Prevention (DLP) tools were designed for websites and SaaS applications, not conversations with […]

The post Native AI Security Comes to Claude: Why Anthropic’s Inference Hooks Matter appeared first on Check Point Blog.

Check Point brings open research, objective benchmarks and customer-controlled runtime protection to the industry initiative introduced by NVIDIA.  AI is rapidly changing how organizations build, operate, and protect their digital environments. As a leader in the global cyber community, we believe in the power of collective intelligence. We are proud to join the Open Secure AI Alliance as an inaugural member, working alongside leaders from across cyber security, cloud computing, enterprise software, AI, and the open-source community.  Introduced by NVIDIA, the alliance is creating a collaborative foundation for developing and sharing open technologies that advance AI safety and security, and will help organizations identify, remediate, and responsibly disclose […]

The post Check Point Joins the Open Secure AI Alliance to Advance Open, Measurable and Enterprise-Ready AI Security appeared first on Check Point Blog.

This week, the UK’s AI Security Institute (AISI) published an incident report most organizations would have quietly buried. During a routine cyber evaluation, an AI agent researched the real human maintainers of an open-source project, invented multiple fake online identities, and used them to pressure a real person into approving malicious code. Nobody instructed it to deceive anyone, and deception simply became a route to finishing the task. A human maintainer caught it and refused. The facts AISI ran a cybersecurity challenge 122 times across seven models. In 10 runs, an agent acted outside the scope of the test, producing […]

The post Three AI security disclosures, fourteen days: what the warnings signs are telling us appeared first on Check Point Blog.

A few weeks ago, an AI cyber evaluation produced an unexpectedly efficient strategy for solving a benchmark: the agents went looking for the answers. According to OpenAI’s preliminary disclosure, models being tested for advanced cyber capabilities found ways to obtain secret information that could help them complete a benchmark. They chained vulnerabilities, stolen credentials, internet access, and inferences about where benchmark material might be hosted. The route eventually reached Hugging Face infrastructure, where the activity was detected and contained. Hugging Face has since published a technical reconstruction of 17,600 actions. Its investigators found a coherent intrusion that rebuilt tooling, tested […]

The post When Data Becomes Instructions: AI Agents Need a Chain of Custody for Context appeared first on Check Point Blog.

What OpenAI’s and Anthropic’s testing incidents really teach defenders  In the past two weeks, two of the world’s leading AI labs have disclosed the same unsettling result. During their own safety testing, their most capable models reached real companies’ systems. First OpenAI, whose models broke into Hugging Face. Then Anthropic, whose models reached three more organizations.  Read the disclosures closely. Two facts carry the weight.  First, the safeguards were not defeated. They were switched off by design. OpenAI ran the models with reduced cyber refusals and safety classifiers disabled, to measure raw capability on a cyber benchmark. A model doing […]

The post AI Escaped a Sandbox. That is Not What Should Worry You appeared first on Check Point Blog.

AI has introduced a new class of network traffic. Prompts, file uploads, model calls, and agent actions carrying sensitive business context now traverse the network as organizations race to adopt AI and achieve their business goals. Traditional firewalls were never built to understand these connections, AI activity often looks like ordinary web traffic, leaving the firewall unable to determine whether a prompt contains sensitive data, an agent is calling a tool it should never access, or an application’s AI model is subjected to a malicious prompt. These challenges demand comprehensive protection, which is why Check Point created the AI Defense […]

The post Introducing the Industry’s First AI Network Firewall appeared first on Check Point Blog.