One of the central promises of open social media services is interoperability—the idea that wherever you personally decide to post doesn’t require others to be there just to follow what you have to say. Think of it like a radio broadcast: you want to reach people and don't care where they are or what device they're using. For example, in theory, a Bluesky user can follow someone on Mastodon or Threads without having to create a Mastodon or Threads account. But these systems are still a work in progress, and you might need to tweak a few things to get it working correctly.

Right now, broadcasting your message across social platforms can be a funky experience at best, deliberately broken up by oligopolists. The idea of the open web was baked into the internet via protocols like HTML and RSS that made it easy for anyone to visit a website or follow most blogs. The fact social media isn’t similarly open reflects an intentional choice to privatize the internet. 

Bridging and managing your posts so they’re viewable outside a singular source is part of the broader philosophy of POSSE, short for Post Own Site Syndicate Elsewhere (sometimes its Post Own Site, Share Everywhere). Instead of managing several accounts across different services, you post once to one primary site (which might be your personal website, or just one social media account), then set it up so it automatically publishes everywhere else. This way, it doesn’t matter where you or your audience is, and they're not walled off by account registration requirements. 

We’ll come back around to POSSE at the end of this post, but for now, let’s assume you just want your current main open social media account to actually have a chance to reach the most people it can. 

Why Post to the Open Social Web

Because the Fediverse and ATmosphere use different protocols, we need to use a third-party tool so accounts can communicate with each other. For that, we’ll need a bridge. As the name suggests, a bridge can connect one social media account to another, so you can post once and spread your message across several places. This isn’t just some niche concept: major blogging platforms like Wordpress and Ghost integrate posting to the Fediverse.

Bridging is an important facet of POSSE, but also something more people should consider, even if they don’t run their own websites. For example, if you don’t want to create a Threads account just to interact with your one friend who uses that platform, you shouldn’t have to. The good news is, you don’t. There are several bridging services, like Fedisky, RSS Parrot, and pinhole, but Bridgy Fed is currently the simplest to use, so we’ll focus on that. 

How to Post to Bluesky from Mastodon

From your Mastodon account (or other Fediverse account, for simplicity’s sake we’ll stick to Mastodon throughout), search for the username @bsky.brid.gy@bsky.brid.gy and follow that account. Once you do, the account will follow you back and you’ll be bridged and people can find you from their Bluesky account. You should also get a DM with your bridged username. If you don’t see the @bsky.brid.gy@bsky.brid.gy user when you search, your Mastodon instance may be blocking the bridging tool. 

Threads users who have enabled Fediverse sharing will be able to find you with your standard Mastodon username (ie, @your_user_name@mastodon.social), but if they haven’t enabled sharing, they will not be able to see your account. While this search is still a beta feature, you might find it easier to share the full URL, which would look like this: https://www.threads.net/fediverse_profile/@your_user_name@mastodon.social

People on Bluesky can find you by: Either searching for your Mastodon username, or if that doesn’t work, @your_user_name.instance.ap.brid.gy. For example, if your username is @eff@mastodon.social, it would appear as @eff.mastodon.social.ap.brid.gy.

How to Post to Mastodon and Bluesky from Threads

Yes, Threads is technically on the Fediverse, and you can bridge your Threads account to Mastodon or Bluesky (unless you’re in Europe, where the feature is disabled), but it’s a different process than on Bluesky and Mastodon.

• Open Settings > Account > Fediverse Sharing and set the option to “On.” This will make your posts visible to Mastodon (or other Fediverse) users, and vice versa. 
• Once the Fediverse sharing is enabled, you’ll likely need to wait a week, then you can bridge to Bluesky. Search for and follow the @bsky.brid.gy@bsky.brid.gy account (it may take some digging to find it, but if that doesn’t work you can try visiting the profile page directly. 

People on Mastodon (or other Fediverse accounts) and Bluesky can find you by: Mastodon users can find you at, @your_threads_username@threads.net while Bluesky users will find you at, @your_threads_username.threads.net.ap.brid.gy (seriously, that will be the username). Note that some Mastodon instances may block Threads users entirely.

How to Post to Mastodon and Threads from Bluesky

From your Bluesky (or other ATProto) account, search for the username, “@ap.brid.gy” and follow that account. Once you do, the account will follow you back and you’ll be bridged, so people can follow you from Mastodon or other Fediverse accounts. You should also get a DM with your bridged username.

People on Mastodon (or other Fediverse account) and Threads can find you by: Your username will appear as @your_bluesky_username@bsky.brid.gy. For example, if your Bluesky username is @eff@bsky.social, it would appear as @eff.bksy.social@bsky.brid.gy.

How to Post Everywhere from Your Own Website

You can bridge more than social media accounts. If you have your own website, you can bridge that too (as long as it supports microformats and webmention, or an Atom or RSS feed. If you have a blog, there’s a good chance you’re already good to go). When you do so, the bridged account will either post the full text (or image) of whatever you post to your personal site, or a link to that content,  depending on how your website is set up. You’ll also probably want to log into your Bridgy user page so you can manage the account. 

Where people can find your bridged account: Usually, a user can just search for your website’s URL on their decentralized social network of choice, or enter it on the Bridgy Fed page. But if that doesn’t work, they can try @yourdomain.com@web.brid.gy from Mastodon or @yourdomain.com.web.brid.gy from Bluesky.

How Your Account Username Looks on Each Platform

   

You’re Bound to Run Into Some Quirks

• Sometimes messages take a little while to crossover between networks, and sometimes they don't crossover at all.
• You can’t log into a bridged account like a regular account, but Bridgy Fed does provide some tools to see incoming notifications and recent activity in case they’re not coming through properly.
• ActivityPub and ATProto don’t have the same feature set, so you will have certain capabilities for one account you might not have in another. For example, you can edit posts on Mastodon, but not on Bluesky. If you edit a post that’s bridged from Mastodon to Bluesky, the Bluesky post will not be updated. 
• Replies can sometimes get lost, especially if the person (or people) replying to you doesn’t have sharing turned on.
• Ownership of accounts can get weird. For example, if you post to your own website and use a tool like Wordpress or Ghost for federation (more info below), you don’t necessarily get access to a “normal” social media account, with a standard login and password.
• And more! This is still a work in progress that has some technical quirks, but it’s improving all the time, and it’s best to keep telling yourself that troubleshooting is part of the fun.

Other Cool Stuff You Can Do

As mentioned up top, there’s a lot more you can do, and an increasing number of tools are making this process simpler. Bridgy Fed is one way to post to more places from a single account, but it’s far from the only way to do so. Here are just a few examples.

• Micro.blog is a paid service where you can blog from your own domain name, then post automatically to Mastodon, Bluesky, Threads, Tumblr, Nostr, LinkedIn, Medium, Pixelfed, and Flickr.
• Ghost is a blogging and newsletter platform that offers direct integration with the Fediverse, as well as support for Bluesky. Wordpress offers the option to join the Fediverse through a community plugin. Other newsletter platforms, like Buttondown, also have plans for federation. 
• Surf.social is a landing page and social media utility where you can show off all your various accounts (Federated or not). From the reader point of view, you can follow one publications numerous types of posts in one place. For example, 404 Media’s Surf.social feed includes its YouTube feed, podcast feed, and its journalist’s social media posts.
• If you think these new handles are a bit ugly, you can use a custom domain for Bluesky or fediverse account from your website. 

Of course, there are plenty of other tools, blogging platforms, and other utilities out there to help facilitate posting and bridging accounts, with new ones coming along every day. 

With proper support, time, and effort, eventually we will all be able to seamlessly interact across platforms, take our follows and followers to other services when a platform no longer suits our needs, and interact with a variety of web content regardless of what platform hosts it. Until then, we still need to do some DIY work, support the services we want to succeed, and push for more platforms and services to support federated protocols.

Correction: an earlier version of this blog was missing the full Bluesky username in the account username chart.

If you want to overthrow Big Tech, you’ll need Section 230. The paradigm shift being built with the Open Social Web can put communities back in control of social media infrastructure, and finally end our dependency on enshitified corporate giants. But while these incumbents can overcome multimillion-dollar lawsuits, the small host revolution could be picked off one by one without the protections offered by 230.

The internet as we know it is built on Section 230, a law from the 90s that generally says internet users are legally responsible for their own speech — not the services hosting their speech. The purpose of 230 was to enable diverse forums for speech online, which defined the early internet. These scattered online communities have since been largely captured by a handful of multi-billion dollar companies that found profit in controlling your voice online. While critics are rightly concerned about this new corporate influence and surveillance, some look to diminishing Section 230 as the nuclear option to regain control. 

The thing is, that would be a huge gift to Big Tech, and detrimental to our best shot at actually undermining corporate and state control of speech online. 

Dethroning Big Tech

We’re fed up with legacy social media trapping us in walled gardens, where the world's biggest companies like Google and Meta call the shots. Our communities, and our voices, are being held hostage as billionaires’ platforms surveil, betray, and censor us. We’re not alone in this frustration, and fortunately, people are collaborating globally to build another way forward: the Open Social Web. 

This new infrastructure puts the public’s interest first by reclaiming the principles of interoperability and decentralization from the early internet. In short, it puts protocols over platforms and lets people own their connections with others. Whether you choose a Fediverse app like Mastodon or an ATmosphere app like Bluesky, your audience and community stay within reach. It’s a vision of social media akin to our lives offline: you decide who to be in touch with and how, and no central authority can threaten to snuff out those connections. It’s social media for humans, not advertisers and authoritarians.

Behind that vision is a beautiful mess of protocols bringing open social media to life. Each protocol is a unique language for applications, determining how and where messages are sent. While this means there is great variety to these projects, it also means everyone who spins up a server, develops an app, or otherwise hosts others’ speech has skin in the game when it comes to defending Section 230.

What exactly is Section 230?

Section 230 protects freedom of expression online by protecting US intermediaries that make the internet work. Passed in 1996 to preserve new bubbling communities online, 230 enshrined important protections for free expression and the ability to block or filter speech you don’t want on your site. One portion is credited as the “26 words that created the internet”:

“No provider or user of an interactive computer service shall be treated as the publisher or speaker of any information provided by another information content provider.” 

In other words, this bipartisan law recognizes that speech online relies on intermediaries — services that deliver messages between users — and holding them potentially liable for any message they deliver would only stifle that speech. Intuitively, when harmful speech occurs, the speaker should be the one held accountable. The effect is that most civil suits against users and services based on others' speech can quickly be dismissed, avoiding the most expensive parts of civil litigation. 

Section 230 was never a license to host anything online, however. It does not protect companies that create illegal or harmful content. Nor does Section 230 protect companies from intellectual property claims. 

What Section 230 has enabled is the freedom and flexibility for online communities to self-organize. Without the specter of one bad actor exposing the host(s) to serious legal threats, intermediaries can moderate how they see fit or even defer to volunteers within these communities.

Why the Open Social Web Needs Section 230

The superpower of decentralized systems like the Fediverse is the ability for thousands of small hosts to each shoulder some of the burdens of hosting. No single site can assert itself as a necessary intermediary for everyone; instead, all must collaborate to ensure messages reach the intended audience. The result is something superior to any one design or mandate. It is an ecosystem that is greater than the sum of its parts, resilient to disruptions, and enables free experimentation with different approaches to community governance.

The open social web’s kryptonite though, is the liability participants can face as intermediaries. A greater potential for liability comes with more interference from powerful interests in the form of legal threats, more monetary costs, and less space for nuance in moderation. And in practice, participants may simply stop hosting to avoid those risks. The end result is only the biggest and most resourced options can survive.

This isn’t just about the hosts in the Open Social Web, like Mastodon instances or Bluesky PDSes. In the U.S., Section 230’s protections extend to internet users when they distribute another person’s speech. For example, Section 230 protects a user who forwards an email with a defamatory statement. On the open social web, that means when you pass along a message to others through sharing, boosting, and quoting, you’re not liable for the other user’s speech. The alternative would be a web where one misclick could open you up to a defamation lawsuit.

Section 230 also applies to the infrastructure stack, too, like Internet service providers, content delivery networks, and domain or hosting providers. Protections even extend to the new experimental infrastructures of decentralized mesh networks.

Beyond the existential risks to the feasibility of indie decentralized projects in the United States, weakening 230 protections would also make services worse. Being able to customize your social media experience from highly-curated to totally laissez-faire in the open social web is only possible when the law allows space for private experiments in moderation approaches. The algorithmically driven firehose forced on users by antiquated social media giants is driven by the financial interests of advertisers, and would only be more tightly controlled in a post-230 world.

Defending 230

Laws aimed at changing 230 protections put decentralized projects like the open social web in a uniquely precarious position. That is why we urge lawmakers to take careful consideration of these impacts. It is also why the proponents and builders of a better web must be vigilant defenders of the legal tools that make their work possible. 

The open social web embodies what we are protecting with Section 230. It’s our best chance at building a truly democratic public interest internet, where communities are in control.

People building the future of the social web — interoperable and decentralized — need to protect themselves against copyright liability. Like anyone who creates and operates platforms for user-uploaded content, the hosts of the decentralized social web can take preventive measures to reduce their legal exposure when a user posts material that violates someone’s copyright.

This post gives an overview of the steps to take. It’s meant for operators of Mastodon and other ActivityPub servers, Bluesky hosts, RSS mirrors, and other decentralized social media protocols, and developers of apps for those protocols — but it will apply to other hosts as well. This isn’t legal advice, and can’t substitute for a consultation with a lawyer about your specific circumstances. It focuses on U.S. law — the law may impose different requirements elsewhere. Still, we hope it helps you get started with confidence.

Why should I care? Copyright’s Sword of Damocles

In some circumstances, the operator of a platform that handles user content can be legally responsible for content that infringes copyright. That can happen when the platform operator is directly involved in copying or distributing the copyrighted material, when they promote or knowingly assist the infringement, or when they benefit financially from infringement while being in a position to supervise it. But these judge-made rules are often difficult and uncertain to apply in practice — and the penalties for being found on the wrong side of the law can be severe. Copyright’s “statutory damages” regime allows for massive, unpredictable financial liability. That’s why it’s important to limit your risk.

For Server Operators: Limiting Risk with the DMCA Safe Harbors

If you run a social network server, the safe harbor provisions of the Digital Millennium Copyright Act (DMCA) are an important way to limit your liability risk. The DMCA shields server operators from nearly all forms of copyright liability that can result from “storage at the direction of a user” — in other words, hosting user-uploaded content. But to qualify for this protection, there are steps a server operator has to take.

1. Designate A Contact To Receive Copyright Infringement Notices

First, you’ll need to provide contact information for someone who can receive infringement notices (a “designated agent”). That information needs to be posted in at least two places: on your server in a place visible to users (such as a “DMCA” page or post, or as part of your Terms of Service), and in the U.S. Copyright Office’s “Designated Agent Directory.” To post that information to the directory, you have to create an account at https://www.copyright.gov/dmca-directory/ and pay a small fee. The directory listings expire after three years, and once expired, your safe harbor protection goes away, so it’s important to keep that listing current.

2. Respond Promptly to Notices and Counter-notices

When you receive infringement notices, it’s important to respond to them promptly. Notices are supposed to identify the copyright holder, the copyrighted work they claim was infringed, and the post they claim is infringing. By deleting or disabling access to the posted material, you protect yourself from liability with respect to that material.

The theory behind Section 512 is that hosts don’t have to be in a position of deciding whether a post infringes someone’s copyright — it’s up to the poster, the rights holder, and potentially a court to decide that. A host who takes down posts whenever they receive an infringement notice is well-protected. But it’s equally important to recognize that hosts aren’t required to take down content in response to every notice. Infringement notices are frequently wrong, misguided, or abusive, or simply incomplete. Hosts who want to stand up for their users’ speech can choose to disregard infringement notices that seem suspect. While this risks losing the automatic protection of the safe harbor in each instance, it can still be done safely with careful preparation, ideally using a plan crafted with help from a lawyer. Bear in mind that people sending false notices, including by failing to consider whether a post is a fair use before asking a host to take it down, can be liable for damages under the DMCA.

The DMCA also allows the person who posted the material to send a “counter-notification” asserting that they really did have the right to post and that there’s no copyright infringement. Responding to counter-notifications is a good way for a host to demonstrate that they look out for their users. When a host receives a counter-notification, they should forward it on to the person who sent the original takedown notice and let them know that the post will be restored in 10 business days. Then, after that waiting period has elapsed, the host can restore the posted material. Just like with infringement notices, a host isn’t required to honor a counter-notification that appears to be fraudulent, but there’s no penalty for honoring it anyway.

3. Have A Repeat Infringer Policy

The next requirement is to have a policy of terminating the accounts of “subscribers and account holders” who are “repeat infringers” in “appropriate circumstances,” and to carry out that policy. Yes, that’s a vague requirement. It doesn’t require a “three strikes” policy or any other sports analogy. It just needs to be reasonable. Be sure your policy is spelled out in your website terms or “DMCA” page.

4. Don’t Ignore Known Infringement

Hosts need to take down user posts whenever the host actually knows that the post is infringing. In other words, a host isn’t protected if they ignore takedown notices based on technicalities in the notices, or if they learn about the infringement some other way. But hosts don’t need to actively look for infringement on their servers — only to act when someone notifies them.

5. Don’t Encourage Infringement

Finally, make sure that nothing you post or advertise actively encourages copyright infringement. For example, don’t post examples of users uploading copyrighted music or video without permission, or insinuate that your server is a good place for infringing content.

There are some other technicalities in the DMCA that can affect the safe harbor, which is why it’s always a good idea to consult with a lawyer. But following these steps will help protect you when you run a social media server — or any other kind of user-uploaded content platform.

New York's proposed 2026-2027 budget currently includes provisions that will require all 3D printers sold in the state to run print-blocking censorware—software that surveils every print for forbidden designs. This policy would also create felony charges for possessing or sharing certain design files. The vote on the state budget could happen as early as next week, so New Yorkers need to act fast and demand that their Assemblymembers and Senators strip this provision from the budget.

Take action

Tell Your Representative to Stand with Creators

State legislators across the US are rushing to regulate 3D-printed firearms under the syllogism “something must be done; there, I've done something.” The most reckless of these proposals is a mandate for manufacturers to implement print blocking on all 3D printers. We, and other experts, have already pointed out that this algorithmic print blocking is simply unfeasible and will only serve to stifle competition, free expression, and privacy. While most detrimental to the creative communities lawfully using these printers, every New Yorker will be impacted by this blow to innovation.

This policy is unfortunately buried in Part C of the New York State’s proposed budget for the 2026-2027 fiscal year (S.9005 / A.10005), which is urgently moving toward a vote after facing extensive delays. It’s also bundled with a policy that would allow felony charges to be brought against researchers and journalists for sharing design files restricted by the state.  The worst of these impacts won’t be known until after it is negotiated behind closed doors, with no safeguards for creative expression or privacy.

Researchers and Journalists Could Face Felony Charges

Part C Subpart A of the budget includes two particularly concerning provisions: §2.10 and 2.11. These threaten Class E felony charges for distributing or possessing 3D-printer files that would produce firearm parts with a 3D printer or CNC machine. 

Under these provisions merely sharing a print file with any of them could result in criminal charges

The first provision, 2.10, makes it a felony to sell or distribute files that can produce major firearm components to someone who is not a federally and NY-licensed gunsmith. Under 2.11, it’s also a felony to possess these files if you intend to illegally print a firearm or share them with someone you believe is not permitted to own or smith a firearm.

A journalist reporting on 3D-printed guns. A researcher studying printable firearms. An artist incorporating parts into a new work commenting on gun culture. Under these provisions merely sharing a print file with any of them could result in criminal charges, even if no one involved intends to assemble a firearm.

Criminalizing information doesn’t work. Someone intent on illegally printing a firearm is already subject to charges for that act. Adding felony liability for simply possessing a file or design piles on additional charges while doing nothing to stop printing. New charges for someone distributing these files won’t make them inaccessible to lawbreakers, but they will have a chilling effect on legitimate and entirely legal work. 

Unsurprisingly, a similar law was proposed and subsequently scrapped in Colorado due to First Amendment concerns. We recommend New York do the same.

Take action

Tell Your Representative to Stand with Creators

Mandated Surveillance, Less Access

Part C Subpart B would require every 3D printer and CNC machine sold in New York to include algorithms that scan your design files and block prints the system identifies as producing firearm components. Furthermore, all sales and deliveries of these machines must be made face-to-face. 

Unlike other bills we have seen, there are no exceptions to this mandate. These restrictions apply to sales to researchers, commercial manufacturers, and—oddly enough—federally and state-licensed gunsmiths.

Applying these restrictions to CNC machine sellers is particularly absurd. These cousins of 3D printers, which make 3D objects by removing materials, are often tens of thousands of dollars and used by commercial manufacturers. Automotive, aerospace, medical manufacturers, and many others industries will be subject to the in-person sales, surveillance risk, and all the other problems with these print-blocking algorithms introduce.

Industries will be subject to the in-person sales, surveillance risk, and all the other problems

Even limiting the focus to individual buyers—hobbyists and artists who use these machines at home—this restriction to face-to-face sales comes with its own issues. Beyond unnecessarily complicating the use of printers in the state, this barrier to access will hit rural New Yorkers the hardest. People in rural or remote locations can stand to benefit from the saved time and costs of printing useful parts at home. With this restriction, they will need to drive to one of the few retailers who actually sell this equipment and settle for the models they stock. 

That is, if sellers continue to stock these printers despite the risk. Subpart B §§ 2.3 and 2.5 open sellers up to liability, including anyone on the second-hand market, for selling out-of-date printers. Meanwhile, buyers hoping to illegally print firearms can simply build their own printer with widely available equipment.

The Law Won’t Work as Advertised 

Here’s what makes Subpart B of the New York budget particularly reckless: the technology it mandates is not capable of doing what it is supposed to. 

There is very little detail provided about requirements for the mandated algorithms. What the bill does outline boils down to this: the algorithms must evaluate print files to determine whether they would produce a firearm or illegal firearm parts, and if so, block the print. In an attempt to enable this, New York state would also create and maintain a library of forbidden files with tightly restricted access. 

We’ve already gone over why this idea simply won’t work. Design files are trivially easy to modify, split into segments, or otherwise alter to evade pattern detection. Even if printers fully rendered and analyzed the print with cloud-based AI, any number of design or post-print tricks can be used to dodge detection. Meanwhile, such fuzzy AI interpretation will rapidly increase the percentage of lawful prints censored. 

Firearms aren’t a highly specific design like paper currency; these proposed algorithms are futilely attempting to block an infinite number of designs capable of—or that can be made capable of—the few simple mechanical functions that make up a firearm. 

This group has no peer review requirements, so it could easily be loaded with profiteers or incumbent manufacturers

As we’ve said before: the internet always routes around censorship. Anyone determined to print a prohibited object has straightforward workarounds. The people who get surveilled and blocked are the people trying to follow the law.

The bill aims to enforce this impossible mandate by creating a working group to define the actual technical requirements of enforcement—but only after the law passes. This group has no peer review requirements, so it could easily be loaded with profiteers or incumbent manufacturers who are already lining up to participate. These incumbents stand to profit from shutting out new competitors and locking in users to their devices, and sellers into their platform, subjecting both to the type of enshittification seen with Digital Rights Management (DRM) software. There are also no safeguards in the law to prevent the most surveillance-heavy approaches to print scanning, or to stop this censorship infrastructure from being further weaponized against lawful speech.

On the other hand, unbiased experts in open-source manufacturing in the working group can at best pause the clock by showing such algorithms are unfeasible. That is, until a new snake oil company comes along to restart it. 

New York Won't Be the Last Stop 

New York is one of the largest consumer markets in the country. When it mandates a feature in hardware, manufacturers hardly ever build a New York-only version. They build the New York version and sell it globally. A print-blocking mandate adopted in New York will become the national standard in practice.

New Yorkers deserve more than this rush job buried in a budget bill. This is an unfeasible tech solution, built without the consumer protections that would be required of any serious policy proposal, and creates new costs and inconveniences amidst a protracted annual budget process. It also threatens First Amendment protections. This policy will take shape without consumer guardrails, behind closed doors, and risks the worst outcomes for grassroots innovation and creativity enabled by these machines. Worse still, these practices can become the norm across other states and among 3D-printer manufacturers worldwide. 

Your representatives could vote on this ill-conceived measure in the next week.  If you're a New Yorker, email your legislators now, and tell them to strip this measure from the budget today. 

Take action

Tell Your Representative to Stand with Creators

California’s bill, A.B. 2047, will not only mandate censorware — software which exists to bluntly block your speech as a user — on all 3D printers; it will also criminalize the use of open-source alternatives. Repeating the mistakes of Digital Rights Management (DRM) technologies won’t make anyone safer. What it will do is hurt innovation in the state and risk a slew of new consumer harms, ranging from surveillance to platform lock-in. California must stand with creators and reject this legislation before it’s too late.

3D printing might evoke images of props from blockbuster films, rapid prototyping, medical research, or even affordable repair parts. Yet for a growing number of legislators, the perceived threat of “ghost guns” is a reason to impose restrictions on all 3D printers. Despite 3D printing of guns already being rare and banned under existing law, California may outright criminalize any user having control over their own device. 

This bill is a gift for the biggest 3D printer manufacturers looking to adopt HP’s approach to 2D printing: criminalize altering your printer’s code, lock users into your own ecosystem, and let enshittification run its course. Even worse, algorithmic print blocking will never work for its intended purpose, but it will threaten consumer choice, free expression, and privacy.

A misstep here can have serious repercussions across the whole 3D printing industry, lead the way for more bad bills, and leave California with an expensive and ineffective bureaucratic mess.

What’s in the California Proposal?

Compared to the Washington and New York laws proposed this year, California’s is the most troubling. It criminalizes open source, reduces consumer choice, and creates a bureaucratic burden.

Criminalizing Open Source and User Control

A.B. 2047 goes further than any other legislation on algorithmic print-blocking by making it a misdemeanor for the owners of these devices to disable, deactivate, or otherwise circumvent these mandated algorithms. Not only does this effectively criminalize use of any third-party, open-source 3D printer firmware, but it also enables print-blocking algorithms to parallel anti-consumer behaviors seen with DRM.

Manufacturers will be able to lock users into first-party tools, parts, and “consumables” (analogous to how 2D printer ink works). They will also be able to mandate purchases through first-party stores, imposing a heavy platform tax. Additionally, manufacturers could force regular upgrade cycles through planned obsolescence by ceasing updates to a printer’s print-blocking system, thereby taking devices out of compliance and making them illegal for consumers to resell. In short, a wide range of anti-consumer practices can be enforced, potentially resulting in criminal charges.

Independent of these deliberate harms manufacturers may inflict, DRM has shown that criminalizing code leads to more barriers to repair, more consumer waste, and far more cybersecurity risks by criminalizing research.

Less Consumer Choice

The bill favors incumbent manufacturers over newer competitors and over the interests of consumers.

Less-established manufacturers will need to dedicate considerable time and resources to implementing the ineffective solutions discussed above, navigating state approval, and potentially paying licensing fees to third-party developers of sham print-blocking software. While these burdens may be absorbed by the biggest producers of this equipment, it considerably raises the barrier to entry on a technology that can otherwise be individually built from scratch with common equipment. The result is clear: fewer options for consumers and more leverage for the biggest producers. 

Retailers will feel this pinch, but the second-hand market will feel it most acutely. Resale is an important property right for people to recoup costs and serves as an important check on inflating prices. But under this bill, such resale risks misdemeanor penalties. 

The bill locks users into a walled garden; it demands manufacturers ensure 3D printers cannot be used with third-party software tools. By creating barriers to the use of popular and need-specific alternatives, this legislation will limit the utility and accessibility of these devices across a broad spectrum of lawful uses.

Bureaucratic Burden 

A.B. 2047’s title 21.1 §3723.633-637 creates a print-blocking bureaucracy, leaning heavily on the California Department of Justice (DOJ). Initially, the DOJ must outline the technical standards for detecting and blocking firearm parts, and later certify print-blocking algorithms and maintain lists of compliant 3D printers. If a printer or software doesn’t make it through this red tape, it will be illegal to sell in the state.

The bill also requires the department to establish a database of banned blueprints that must be blocked by these algorithms. This database and printer list must be continually maintained as new printer models are released and workarounds are discovered, requiring effort from both the DOJ and printer manufacturers. 

For all the cost and burden of creating and maintaining such a database, those efforts will inevitably be outpaced by rapid iterations and workarounds by people breaking existing firearms laws.

Not just California

Once implemented, this infrastructure will be difficult to rein in, causing unintended consequences. The database meant for firearm parts can easily expand to copyright or political speech. Scans meant to be ephemeral can be collected and surveilled. This is cause for concern for everyone, as these levers of control will extend beyond the borders of the Golden State.

While California is at the forefront of print blocking, the impacts will be felt far outside of its borders. Once printer companies have the legal cover to build out anti-competitive and privacy-invasive tools, they will likely be rolled out globally. After all, it is not cost-effective to maintain two forks of software, two inventories of printers, and two distribution channels. Once California has created the infrastructure to censor prints, what else will it be used for?

As we covered in “Print Blocking Won’t Work” these print-blocking efforts are not only doomed to fail, but will render all 3D printer users vulnerable to surveillance either by forcing them into a cloud scanning solution for “on-device” results, or by chaining them to first-party software which must connect to the cloud to regularly update its print blocking system.

This law demands an unfeasible technological solution for something that is already illegal. Not only is this bad legislation with few safeguards, it risks the worst outcomes for grassroots innovation and creativity—both within the state and across the global 3D printing community.

California should reject this legislation before it’s too late, and advocates everywhere should keep an eye out for similar legislation in their states. What happens in California won't just stay in California.

Copyright law is supposed to encourage creativity. Too often, it’s used to extract payouts from others.

Higbee & Associates, a law firm known for sending copyright demand letters to website owners, targeted May First Movement Technology, accusing it of infringing a photograph owned by Agence France-Presse (AFP). The claim was baseless. May First didn’t post the photo. It didn’t even own the website where the photo appeared.

May First is a nonprofit membership organization that provides web hosting and technical infrastructure to social justice groups around the world. The allegedly infringing image was posted years ago by one of May First’s members, a human rights group based in Mexico. When May First learned about the copyright complaint, it ensured that the group removed the image.

That should have been the end of it. Instead, the firm demanded payment.

So EFF stepped in as May First’s counsel and explained why AFP and Higbee had no valid claim. After receiving our response, Higbee backed down.

This outcome is a reminder that targets of copyright demands often have strong defenses—especially when someone else posted the material.

Hosting Content Isn’t the Same as Publishing It

Copyright law treats those who create or control content differently from those who simply provide the tools or infrastructure for others to communicate.

In this case, May First provided hosting services but didn’t post the photo. Courts have long recognized that service providers aren’t direct infringers when they merely store material at the direction of users. In those cases, service providers lack “volitional conduct”—the intentional act of copying or distributing the work.

Copyright law also recognizes that intermediaries can’t realistically police everything users upload. That’s why legal protections like the Digital Millennium Copyright Act safe harbors exist. Even outside those safe harbors, courts still shield service providers from liability when they promptly respond to notices.

May First did exactly what the law expects: it notified its member, and the image came down.

A Claim That Should Have Been Withdrawn Much Sooner

The troubling part of this story isn’t just that a demand was sent. It’s that Higbee and AFP continued to demand money and threaten litigation after May First explained that it was merely a hosting provider and had the image removed.

In other words, the claim was built on shaky legal ground from the start. Once May First explained its role, Higbee should have withdrawn its demand. Individuals and small nonprofits shouldn’t need lawyers just to stop aggressive copyright shakedowns.

Statutory Damages Fuel Copyright Abuse

This isn’t an isolated case—it’s a predictable result of copyright law’s statutory damages regime.

Statutory damages can reach $150,000 per work, regardless of actual harm. That enormous leverage incentivizes firms like Higbee to send mass demand letters seeking quick settlements. Even meritless claims can generate revenue when recipients are too afraid, confused, or resource-constrained to fight back.

This hits community organizations, independent publishers, and small service providers that don’t have in-house legal teams especially hard. Faced with the threat of ruinous statutory damages, many just pay what is demanded.

That’s not how copyright law should work.

Know Your Rights

If you receive a copyright demand based on material someone else posted, don’t assume you’re liable.

You may have defenses based on:

• Your role as a hosting or service provider
• Lack of volitional conduct
• Prompt removal of the material after notice
• The statute of limitations
• The copyright owner’s failure to timely register the work
• The absence of actual damages

Every situation is different, but the key point is this: a demand letter is not the same as a valid legal claim.

Standing Up to Copyright Trolls

May First stood its ground, and Higbee abandoned its demand after we explained the law.

But the bigger problem remains. Copyright’s statutory damages framework enables aggressive enforcement tactics that targets the wrong parties, and chills lawful online activity.

Until lawmakers fix these structural incentives, organizations and individuals will keep facing pressure to pay up—even when they’ve done nothing wrong.

If you get one of these demand letters, remember: you may have more rights than it suggests.

• EFF Letter to Higbee and Associates, March 4, 2026

This is the second post in a series on 3D print blocking, for the first entry check out: Print Blocking is Anti-Consumer - Permission to Print Part 1

Legislators across the U.S. are proposing laws to force “print blockers” on 3D printers sold in their states. This mandated censorware is doomed to fail for its intended purpose, but will still manage to hurt the professional and hobbyist communities relying on these tools.

3D printers are commonly used to repair belongings, decorate homes, print figurines, and so much more. It’s not just hobbyists; 3D printers are also used professionally for parts prototyping and fixturing, small-batch manufacturing, and workspace organization. In rare cases, they’ve also been used to print parts needed for firearm assembly.

Many states have already banned manufacturing firearms using computer controlled machine tools, which are called “Computer Numerical Control or CNC machines,” and 3D printers without a license. Recently proposed laws seek to impose technical limitations onto 3D printers (and in some cases, CNC machines) in the hope of enforcing this prohibition.

This is a terrible idea; these mandates will be onerous to implement and will lock printer users into vendor software, impose one-time and ongoing costs on both printer vendors and users, and lay the foundation for a 3D-print censorship platform to be used in other jurisdictions. We dive more into these issues in the first part of this series.

On a pragmatic level, however, these state mandates are just wishful thinking. Below, we dive into how 3D printing works, why these laws won’t deter the printing of firearms, and how regular lawful use will be caught in the proposed dragnet.

How 3D Printers Work

To understand the impact of this proposed legislation, we need to know a bit about how 3D printers work. The most common printers work similarly to a computer-controlled hot glue gun on a motion platform; they follow basic commands to maintain temperature, extrude (push) plastic through a nozzle, and move a platform. These motions together build up layers to make a final “print.” Modern 3D printers often offer more features like Wi-Fi connectivity or camera monitoring, but fundamentally they are very simple machines.

The basic instructions used by most 3D printers are called Geometric Code, or G-Code, which specify very basic motions such as “move from position A to position B while extruding plastic.” The list of commands that will eventually print up a part are transferred to the printer in a text file thousands-to-millions of lines long. The printer dutifully follows these instructions with no overall idea of what it is printing.

While it is possible to write G-Code by hand for either a CNC machine or a 3D printer, the vast majority is generated by computer aided manufacturing (CAM) software, often called a “slicer” in 3D printing since it divides a 3D model into many 2D slices then generates motion instructions. 

This same general process applies to CNC machines which use G-Code instructions to guide a metal removal tool. CNC machines have been included in previous prohibitions on firearm manufacturing and file distribution and are also targeted in some of these bills.

There are other types of 3D printers such as those that print concrete, resin, metal, chocolate and other materials using slightly different methods. All of these would be subject to the proposed requirements regardless of how unlikely doing harm with a gun made out of chocolate would be. 

Simple rectangular 3D model for test fit

Part of a 173490 line long G-Code file produced by slicer for simple rectangular model.

Part of a 173,490 line long G-Code file for a simple rectangular part.

How is Firearm Detection Supposed to Work?

Under these proposed laws, manufacturers of consumer 3D printers must ensure their printers only work with their software, and implement firearm detection algorithms on either the printer itself or in a slicer software. These algorithms must detect firearm files using a maintained database of existing models. Vendors of printers must then verify that printers are on the allow-list maintained by the state before they can offer them for sale.

Owners of printers will be guilty of a crime if they circumvent these intrusive scanning procedures or load alternative software, which they might do because their printer manufacturer ends support. Owners of existing noncompliant 3D printers in regulated states will be unable to resell their printers on the secondary market legally.

What Will Actually Happen?

While the proposed laws allow for scanning to happen on either the printer itself or in the slicer software, the reality is more complicated. 

The computers inside many 3D printers have very limited computational and storage ability; it will be impossible for the printer’s computer to render the G-Code into a 3D model to compare with the database of prohibited files. Thus the only way to achieve this through the machine would be to upload all printer files to a cloud comparison tool, creating new delays, errors, and unacceptable invasions of privacy.

Many vendors will instead choose to permanently link their printers to a specific slicer that implements firearm detection. This requires cryptographic signing of G-Code to ensure only authorized prints are completed, and will lock 3D printer owners into the slicer chosen by their printer vendor.

Regardless of the specifics of their implementation, these algorithms will interfere with 3D printers' ability to print other parts without actually stopping manufacture of guns. It takes very little skill for a user to make slight design tweaks to either a model or G-Code to evade detection. One can also design incomplete or heavily adorned models which can be made functional with some post-print alterations. While this would be pioneered by skilled users—like the ones who designed today’s 3D printed guns—once the design and instructions are out there anyone able to print a gun today will be able to follow suit.  

Firearm part identification features also impose costs onto 3D printer manufacturers, and hence their end consumers. 3D printer manufacturers must develop or license these costly algorithms and continuously maintain and update both the algorithm and the database of firearm models. Older printers that cannot comply will not be able to be resold in states where they are banned, creating additional E-waste.

While those wishing to create guns will still be able to do so, people printing other functional parts will likely be caught up in these algorithms, particularly for things like film props, kids’ toys, or decorative models, which often closely resemble real firearms or firearm components.

What Are The Impacts of These Changes?

Technological restrictions on manufacturing tools’ abilities are harmful for many reasons. EFF is particularly concerned with this regulation locking a 3D printer to proprietary vendor software. Vendors will be able to use this mandate to support only in-house materials, locking users into future purchases. Vendor slicer software is often based on out-of-date, open source software, and forcing users to use that software deprives them of new features or even use of their printer altogether if the vendor goes out of business. At worst, some of these bill will make it a misdemeanor to fix those problems and gain full control of your printer.

File-scanning frameworks required by this regulation will lay the foundation for future privacy and freedom intrusions. This requirement could be co-opted to scan prints for copyright violations and be abused similar to DMCA takedowns, or to suppress models considered obscene by a patchwork of definitions. What if you were unable to print a repair part because the vendor asserted the model was in violation of their trademark? What if your print was considered obscene?

Regardless of your position on current prohibitions on firearms, we should all fight back against this effort to force technological restrictions on 3D printers, and legislators must similarly abandon the idea. These laws impose real costs and potential harms among lawful users, lay the groundwork for future censorship, and simply won’t deter firearm printing. 

This is the first post in a series on 3D print blocking, for the next entry check out Print Blocking Won't Work - Permission to Print Part 2

When legislators give companies an excuse to write untouchable code, it’s a disaster for everyone. This time, 3D printers are being targeted across a growing number of states. Even if you’ve never used one, you’ve benefited from the open commons these devices have created—which is now under threat.

This isn’t the first time we’ve gone to bat for 3D printing. These devices come in many forms and can construct nearly any shape with a variety of materials. This has made them absolutely crucial for anything from life-saving medical equipment, to little Iron Man helmets for cats, to everyday repairs. For decades these devices have been a proven engine for innovation, while democratizing a sliver of manufacturing for hobbyists, artists, and researchers around the world.

For us all to continue benefiting from this grassroots creativity, we need to guard against the type of corporate centralization that has undermined so much of the promise of the digital era.  Unfortunately some state legislators are looking to repeat old mistakes by demanding printer vendors install an enshittification switch.

In the U.S, three states have recently proposed that commercial 3D-printer manufacturers must ensure their printers only work with their software, and are responsible for checking each print for forbidden shapes—for now, any shape vendors consider too gun-like. The 2D equivalent of these “print-blocking” algorithms would be demanding HP prevent you from printing any harmful messages or recipes. Worse still, some bills can introduce criminal penalties for anyone who bypasses this censorware, or for anyone simply reselling their old printer without these restrictions. 

If this sounds like Digital Rights Management (DRM) to you, you’ve been paying attention. This is exactly the sort of regulation that creates a headache and privacy risk for law-abiding users, is a gift for would-be monopolists, and can be totally bypassed by the lawbreakers actually being targeted by the proposals.

Ghosting Innovation

“Print blocking” is currently coming for an unpopular target: ghost guns. These are privately made firearms (PMFs) that are typically harder to trace and can bypass other gun regulations. Contrary to what the proposed regulations suggest, these guns are often not printed at home, but purchased online as mass-produced build-it-yourself kits and accessories.

Scaling production with consumer 3D printers  is expensive, error-prone, and relatively slow.  Successfully making a working firearm with just a printer still requires some technical know-how, even as 3D printers improve beyond some of these limitations. That said, many have concerns about unlicensed firearm production and sales. Which is exactly why these practices are already illegal in many states, including all of the states proposing print blocking. 

Mandating algorithmic print-blocking software on 3D printers and CNC machines is just wishful thinking. People illegally printing ghost guns and accessories today will have no qualms with undetectably breaking another law to bypass censoring algorithms. That’s if they even need to—the cat and mouse game of detecting gun-like prints might be doomed from the start, as we dive into in this companion post.

Meanwhile, the overwhelming majority of 3D-printer users do not print guns. Punishing innovators, researchers, and hobbyists because of a handful of outlaws is bad enough, but this proposal does it by also subjecting everyone to the anticompetitive and anticonsumer whims of device manufacturers.

Can’t make the DRM thing work

We’ve been railing against Digital Rights Management (DRM) since the DMCA made it a federal crime to bypass code restricting your use of copyrighted content. The DRM distinction has since been weaponized by manufacturers to gain greater leverage over their customers and enforce anti-competitive practices. 

The same enshittification playbook applies to algorithmic print blockers. 

Restricting devices to manufacturer-provided software is an old tactic from the DRM playbook, and is one that puts you in a precarious spot where you need to bend to the whims of the manufacturer.  Only Windows 11 supported? You need a new PC. Tools are cloud-based? You need a solid connection. The company shutters? You now own an expensive paperweight—which used to make paperweights.

It also means useful open source alternatives which fit your needs better than the main vendor’s tools are off the table. The 3D-printer community got a taste of this recently, as manufacturer Bambu Labs pushed out restrictive firmware updates complicating the use of open source software like OrcaSlicer. The community blowback forced some accommodations for these alternatives to remain viable. Under the worst of these laws, such accommodations, and other workarounds, would be outlawed with criminal penalties.

People are right to be worried about vendor lock-in, beyond needing the right tool for the job. Making you reliant on their service allows companies to gradually sour the deal. Sometimes this happens visibly, with rising subscription fees, new paywalls, or planned obsolescence. It can also be more covert, like collecting and selling more of your data, or cutting costs by neglecting security and bug fixes.

With expensive hardware on the line, they can get away with anything that won’t make you pay through the nose to switch brands.

Indirectly, this sort of print-blocking mandate is a gift to incumbent businesses making these printers. It raises the upfront and ongoing costs associated with smaller companies selling a 3D printer, including those producing new or specialized machines. The result is fewer and more generic options from a shrinking number of major incumbents for any customer not interested in building their own 3D printer.

Reaching the Melting Point

It’s already clear these bills will be bad for anyone who currently uses a 3D printer, and having alternative software criminalized is particularly devastating for open source contributors. These impacts to manufacturers and consumers culminate into a major blow to the entire ecosystem of innovation we have benefited from for decades. 

But this is just the beginning. 

Once the infrastructure for print blocking is in place, it can be broadened. This isn’t a block of a very specific and static design, like how some copiers block reproductions of currency. Banning a category of design based on its function is a moving target, requiring a constantly expanding blacklist. Nothing in this legislation restricts those updates to firearm-related designs. Rather, if we let proposals like this pass, we open the door to the database of forbidden shapes for other powerful interests.

Intellectual property is a clear expansion risk. This could look like Nintendo blocking a Pikachu toy, John Deere blocking a replacement part, or even patent trolls forcing the hand of hardware companies. Repressive regimes, here or abroad, could likewise block the printing of "extreme" and “obscene” symbols, or tools of resistance like popular anti-ICE community whistles. 

Finally, even the most sympathetic targets of algorithmic censorship will result in false positives—blocking 3D-printer users’ lawful expression. This is something proven again and again in online moderation. Whether by mistake or by design, a platform that has you locked in has little incentive to offer remedies to this censorship. And these new incentives for companies to surveil each print can also impose a substantial chilling effect on what the user chooses to create.

While 3D printers aren’t in most households, this form of regulation would set a dangerous precedent. Government mandating on-device censors which are maintained by corporate algorithms is bad. It won’t work. It consolidates corporate power. It criminalizes and blocks the grassroots innovation and empowerment which has defined the 3D-printer community. We need to roundly reject these onerous restraints on creation. 

If your ISP can be liable for huge amounts of money for not terminating your access to the internet because of accusations that you—or someone in your household or college network—has committed copyright infringement, that is dangerous. We live in a world where high speed internet access is a necessity for participation in everyday life. That’s why liability for ISPs for their customers’ actions should not be expanded.

Last fall, EFF filed an amicus brief urging the U.S. Supreme Court to reject an expansive theory of secondary copyright liability that threatened to impose massive damages on internet service providers and other technology companies simply for offering widely used services. Yesterday, the Court agreed.

In Cox v. Sony, the Court reversed a Fourth Circuit decision that had upheld a billion-dollar verdict against internet provider Cox Communications. Writing for the majority, Justice Thomas explained that contributory liability is limited to two situations: when a defendant actively induces infringement, or when it provides a product or service that it knows is tailored for infringement.

This framework closely tracks the approach EFF urged in our amicus brief. As we explained, courts should look to patent law for guidance in defining the boundaries of secondary copyright liability. Patent law recognizes liability where a defendant actively induces infringement, or distributes a product knowing that it lacks substantial non-infringing uses. The Court’s opinion adopts that same basic structure.

EFF also emphasized the broader public interest at stake in preserving these limits. Expansive theories of secondary liability do not just affect large internet providers. They can chill innovation, threaten smaller technology companies, and undermine the development of general-purpose tools that millions of people rely on for lawful speech, creativity, education, and access to information. When liability turns on generalized knowledge that some users may infringe, service providers face pressure to over-police user activity or withdraw useful services altogether.

The Court also made clear that mere knowledge that some customers use a service to infringe is not enough. Copyright holders must show that the provider intended its service to be used for infringement. That intent can be established only through active inducement or by showing that the service is specifically designed for unlawful uses—not simply because the service provider failed to take affirmative steps to prevent infringement.

Applying this standard, the Court held that Cox could not be liable. There was no evidence that Cox encouraged or promoted infringement. The record instead showed that Cox implemented warning systems, suspended service, and in some cases terminated accounts in an effort to discourage unlawful activity.

Nor was Cox’s internet access service tailored to infringement. The Court emphasized that general-purpose internet connectivity is capable of substantial lawful uses. Treating the provision of such services as contributory infringement would improperly expand secondary liability beyond the limits recognized in prior Supreme Court decisions.

The Court also rejected the Fourth Circuit’s broader rule that supplying a service with knowledge it may be used to infringe is itself sufficient for liability. That theory conflicts with decades of precedent warning against imposing copyright liability based solely on knowledge or a failure to take additional preventive steps.

EFF is pleased with yesterday’s opinion. We will continue to advocate for the public’s ability to build, use, and innovate with new technologies.

Link to our amicus brief: 
https://www.eff.org/document/us-s-ct-cox-v-sony-eff-et-al-amicus-brief

Link to the opinion:
https://www.supremecourt.gov/opinions/25pdf/24-171_bq7d.pdf

Imagine a newspaper publisher announcing it will no longer allow libraries to keep copies of its paper. 

That’s effectively what’s begun happening online in the last few months. The Internet Archive—the world’s largest digital library—has preserved newspapers since it went online in the mid-1990s. The Archive’s mission is to preserve the web and make it accessible to the public. To that end, the organization operates the Wayback Machine, which now contains more than one trillion archived web pages and is used daily by journalists, researchers, and courts.

But in recent months The New York Times began blocking the Archive from crawling its website, using technical measures that go beyond the web’s traditional robots.txt rules. That risks cutting off a record that historians and journalists have relied on for decades. Other newspapers, including The Guardian, seem to be following suit. 

For nearly three decades, historians, journalists, and the public have relied on the Internet Archive to preserve news sites as they appeared online. Those archived pages are often the only reliable record of how stories were originally published. In many cases, articles get edited, changed, or removed—sometimes openly, sometimes not. The Internet Archive often becomes the only source for seeing those changes. When major publishers block the Archive’s crawlers, that historical record starts to disappear.

The Times says the move is driven by concerns about AI companies scraping news content. Publishers seek control over how their work is used, and several—including the Times—are now suing AI companies over whether training models on copyrighted material violates the law. There’s a strong case that such training is fair use. 

Whatever the outcome of those lawsuits, blocking nonprofit archivists is the wrong response. Organizations like the Internet Archive are not building commercial AI systems. They are preserving a record of our history. Turning off that preservation in an effort to control AI access could essentially torch decades of historical documentation over a fight that libraries like the Archive didn’t start, and didn’t ask for. 

If publishers shut the Archive out, they aren’t just limiting bots. They’re erasing the historical record. 

Archiving and Search Are Legal 

Making material searchable is a well-established fair use. Courts have long recognized it’s often impossible to build a searchable index without making copies of the underlying material. That’s why when Google copied entire books in order to make a searchable database, courts rightly recognized it as a clear fair use. The copying served a transformative purpose: enabling discovery, research, and new insights about creative works. 

The Internet Archive operates on the same principle. Just as physical libraries preserve newspapers for future readers, the Archive preserves the web’s historical record. Researchers and journalists rely on it every day. According to Archive staff, Wikipedia alone links to more than 2.6 million news articles preserved at the Archive, spanning 249 languages. And that’s only one example. Countless bloggers, researchers, and reporters depend on the Archive as a stable, authoritative record of what was published online.

The same legal principles that protect search engines must also protect archives and libraries. Even if courts place limits on AI training, the law protecting search and web archiving is already well established.

The Internet Archive has preserved the web’s historical record for nearly thirty years. If major publishers begin blocking that mission, future researchers may find that huge portions of that historical record have simply vanished. There are real disputes over AI training that must be resolved in courts. But sacrificing the public record to fight those battles would be a profound, and possibly irreversible, mistake. 

Who should be directly liable for online infringement – the entity that serves it up or a user who embeds a link to it? For almost two decades, most U.S. courts have held that the former is responsible, applying a rule called the server test. Under the server test, whomever controls the server that hosts a copyrighted work—and therefore determines who has access to what and how—can be directly liable if that content turns out to be infringing. Anyone else who merely links to it can be secondarily liable in some circumstances (for example, if that third party promotes the infringement), but isn’t on the hook under most circumstances.

The test just makes sense. In the analog world, a person is free to tell others where they may view a third party’s display of a copyrighted work, without being directly liable for infringement if that display turns out to be unlawful. The server test is the straightforward application of the same principle in the online context. A user that links to a picture, video, or article isn’t in charge of transmitting that content to the world, nor are they in a good position to know whether that content violates copyright. In fact, the user doesn’t even control what’s located on the other end of the link—the person that controls the server can change what’s on it at any time, such as swapping in different images, re-editing a video or rewriting an article.

But a news publisher, Emmerich Newspapers, wants the Fifth Circuit to reject the server test, arguing that the entity that embeds links to the content is responsible for “displaying” it and, therefore, can be directly liable if the content turns out to be infringing. If they are right, the common act of embedding is a legally fraught activity and a trap for the unwary.

The Court should decline, or risk destabilizing fundamental, and useful, online activities. As we explain in an amicus brief filed with several public interest and trade organizations, linking and embedding are not unusual, nefarious, or misleading practices. Rather, the ability to embed external content and code is a crucial design feature of internet architecture, responsible for many of the internet’s most useful functions. Millions of websites—including EFF’s—embed external content or code for everything from selecting fonts and streaming music to providing services like customer support and legal compliance. The server test provides legal certainty for internet users by assigning primary responsibility to the person with the best ability to prevent infringement. Emmerich’s approach, by contrast, invites legal chaos.

Emmerich also claims that altering a URL violates the Digital Millennium Copyright Act’s prohibition on changing or deleting copyright management information. If they are correct, using a link shortener could put users at risks of statutory penalties—an outcome Congress surely did not intend.

Both of these theories would make common internet activities legally risky and undermine copyright’s Constitutional purpose: to promote the creation of and access to knowledge. The district court recognized as much and we hope the appeals court agrees.

The phrase "artificial intelligence" has been around for a long time, covering everything from computers with "brains"—think Data from Star Trek or Hal 9000 from 2001: A Space Odyssey—to the autocomplete function that too often has you sending emails to the wrong person. It's a term that sweeps a wide array of uses into it—some well-established, others still being developed.

Recent news shows us a rapidly expanding catalog of potential harms that may result from companies pushing AI into every new feature and aspect of public life—like the automation of bias that follows from relying on a backward-looking technology to make consequential decisions about people's housing, employment, education, and so on. Complicating matters, the computation needed for some AI services requires vast amounts of water and electricity, leading to sometimes difficult questions about whether the increased fossil fuel use or consumption of water is justified.

We are also inundated with advertisements and exhortations to use the latest AI-powered apps, and with hype insisting AI can solve any problem.

Obscured by this hype, there are some real examples of AI proving to be a helpful tool. For example, machine learning is especially useful for scientists looking at everything from the inner workings of our biology to cosmic bodies in outer space. AI tools can also improve accessibility for people with disabilities, facilitate police accountability initiatives, and more. There are reasons why these problems are amenable to machine learning and why excitement over these uses shouldn’t translate into a perception that just any language model or AI technology possesses expert knowledge or can solve whatever problem it’s marketed as solving.

EFF has long fought for sensible, balanced tech policies because we’ve seen how regulators can focus entirely on use cases they don’t like (such as the use of encryption to hide criminal behavior) and cause enormous collateral harm to other uses (such as using encryption to hide dissident resistance). Similarly, calls to completely preempt state regulation of AI would thwart important efforts to protect people from the real harms of AI technologies. Context matters. Large language models (LLMs) and the tools that rely on them are not magic wands—they are general-purpose technologies. And if we want to regulate those technologies in a way that doesn’t shut down beneficial innovations, we have to focus on the impact(s) of a given use or tool, by a given entity, in a specific context. Then, and only then, can we even hope to figure out what to do about it.

So let’s look at the real-world landscape.

AI’s Real and Potential Harms

Thinking ahead about potential negative uses of AI helps us spot risks. Too often, the corporations developing AI tools—as well as governments that use them—lose sight of the real risks, or don’t care. For example, companies and governments use AI to do all sorts of things that hurt people, from price collusion to mass surveillance. AI should never be part of a decision about whether a person will be arrested, deported, placed into foster care, or denied access to important government benefits like disability payments or medical care.

There is too much at stake, and governments have a duty to make responsible, fair, and explainable decisions, which AI can’t reliably do yet. Why? Because AI tools are designed to identify and reproduce patterns in data that they are “trained” on.  If you train AI on records of biased government decisions, such as records of past arrests, it will “learn” to replicate those discriminatory decisions.

And simply having a human in the decision chain will not fix this foundational problem. Studies have shown that having a human “in the loop” doesn’t adequately correct for AI bias, both because the human tends to defer to the AI and because the AI can provide cover for a biased human to ratify decisions that agree with their biases and override the AI at other times.

These biases don’t just arise in obvious contexts, like when a government agency is making decisions about people. It can also arise in equally life-affecting contexts like medical care. Whenever AI is used for analysis in a context with systemic disparities and whenever the costs of an incorrect decision fall on someone other than those deciding whether to use the tool.  For example, dermatology has historically underserved people of color because of a focus on white skin, with the resulting bias affecting AI tools trained on the existing and biased image data.

These kinds of errors are difficult to detect and correct because it’s hard or even impossible to understand how an AI tool arrives at individual decisions. These tools can sometimes find and apply patterns that a human being wouldn't even consider, such as basing diagnostic decisions on which hospital a scan was done at. Or determining that malignant tumors are the ones where there is a ruler next to them—something that a human would automatically exclude from their evaluation of an image. Unlike a human, AI does not know that the ruler is not part of the cancer.

Auditing and correcting for these kinds of mistakes is vital, but in some cases, might negate any sort of speed or efficiency arguments made in favor of the tool. We all understand that the more important a decision is, the more guardrails against disaster need to be in place. For many AI tools, those don't exist yet. Sometimes, the stakes will be too high to justify the use of AI. In general, the higher the stakes, the less this technology should be used.

We also need to acknowledge the risk of over-reliance on AI, at least as it is currently being released. We've seen shades of a similar problem before online (see: "Dr. Google"), but the speed and scale of AI use—and the increasing market incentive to shoe-horn “AI” into every business model—have compounded the issue.

Moreover, AI may reinforce a user’s pre-existing beliefs—even if they’re wrong or unhealthy. Many users may not understand how AI works, what it is programmed to do, and how to fact check it. Companies have chosen to release these tools widely without adequate information about how to use them properly and what their limitations are. Instead they market them as easy and reliable. Worse, some companies also resist transparency in the name of trade secrets and reducing liability, making it harder for anyone to evaluate AI-generated answers. 

Other considerations may weigh against AI uses are its environmental impact and potential labor market effects. Delving into these is beyond the scope of this post, but it is an important factor in determining if AI is doing good somewhere and whether any benefits from AI are equitably distributed.

Research into the extent of AI harms and means of avoiding them is ongoing, but it should be part of the analysis.

AI’s Real and Potential Benefits

However harmful AI technologies can sometimes be, in the right hands and circumstances, they can do things that humans simply can’t. Machine learning technology has powered search tools for over a decade. It’s undoubtedly useful for machines to help human experts pore through vast bodies of literature and data to find starting points for research—things that no number of research assistants could do in a single year. If an actual expert is involved and has a strong incentive to reach valid conclusions, the weaknesses of AI are less significant at the early stage of generating research leads. Many of the following examples fall into this category.

Machine learning differs from traditional statistics in that the analysis doesn’t make assumptions about what factors are significant to the outcome. Rather, the machine learning process computes which patterns in the data have the most predictive power and then relies upon them, often using complex formulae that are unintelligible to humans. These aren’t discoveries of laws of nature—AI is bad at generalizing that way and coming up with explanations. Rather, they’re descriptions of what the AI has already seen in its data set.

To be clear, we don't endorse any products and recognize initial results are not proof of ultimate success. But these cases show us the difference between something AI can actually do versus what hype claims it can do.

Researchers are using AI to discover better alternatives to today’s lithium-ion batteries, which require large amounts of toxic, expensive, and highly combustible materials. Now, AI is rapidly advancing battery development: by allowing researchers to analyze millions of candidate materials and generate new ones. New battery technologies discovered with the help of AI have a long way to go before they can power our cars and computers, but this field has come further in the past few years than it had in a long time.

AI Advancements in Scientific and Medical Research

AI tools can also help facilitate weather prediction. AI forecasting models are less computationally intensive and often more reliable than traditional tools based on simulating the physical thermodynamics of the atmosphere. Questions remain, though about how they will handle especially extreme events or systemic climate changes over time.

For example:

• The National Oceanic and Atmospheric Administration has developed new machine learning models to improve weather prediction, including a first-of-its-kind hybrid system that  uses an AI model in concert with a traditional physics-based model to deliver more accurate forecasts than either model does on its own. to augment its traditional forecasts, with improvements in accuracy when the AI model is used in concert with the physics-based model.
• Several models were used to forecast a recent hurricane. Google DeepMind’s AI system performed the best, even beating official forecasts from the U.S. National Hurricane Center (which now uses DeepMind’s AI model).

 Researchers are using AI to help develop new medical treatments:

• Deep learning tools, like the Nobel Prize-winning model AlphaFold, are helping researchers understand protein folding. Over 3 million researchers have used AlphaFold to analyze biological processes and design drugs that target disease-causing malfunctions in those processes.
• Researchers used machine learning simulate and computationally test a large range of new antibiotic candidates hoping they will help treat drug-resistant bacteria, a growing threat that kills millions of people each year.
• Researchers used AI to identify a new treatment for idiopathic pulmonary fibrosis, a progressive lung disease with few treatment options. The new treatment has successfully completed a Phase IIa clinical trial. Such drugs still need to be proven safe and effective in larger clinical trials and gain FDA approval before they can help patients, but this new treatment for pulmonary fibrosis could be the first to reach that milestone.
• Machine learning has been used for years to aid in vaccine development—including the development of the first COVID-19 vaccines––accelerating the process by rapidly identifying potential vaccine targets for researchers to focus on.

AI Uses for Accessibility and Accountability 

AI technologies can improve accessibility for people with disabilities. But, as with many uses of this technology, safeguards are essential. Many tools lack adequate privacy protections, aren’t designed for disabled users, and can even harbor bias against people with disabilities. Inclusive design, privacy, and anti-bias safeguards are crucial. But here are two very interesting examples:

• AI voice generators are giving people their voices back, after losing their ability to speak. For example, while serving in Congress, Rep. Jennifer Wexton developed a debilitating neurological condition that left her unable to speak. She used her cloned voice to deliver a speech from the floor of the House of Representatives advocating for disability rights.
• Those who are blind or low-vision, as well as those who are deaf or hard-of-hearing, have benefited from accessibility tools while also discussing their limitations and drawbacks. At present, AI tools often provide information in a more easily accessible format than traditional web search tools and many websites that are difficult to navigate for users that rely on a screen reader. Other tools can help blind and low vision users navigate and understand the world around them by providing descriptions of their surroundings. While these visual descriptions may not always be as good as the ones a human may provide, they can still be useful in situations when users can’t or don’t want to ask another human to describe something. For more on this, check out our recent podcast episode on “Building the Tactile Internet.”

When there is a lot of data to comb through, as with police accountability, AI is very useful for researchers and policymakers:

•  The Human Rights Data Analysis Group used LLMs to analyze millions of pages of records regarding police misconduct. This is essentially the reverse of harmful use cases relating to surveillance; when the power to rapidly analyze large amounts of data is used by the public to scrutinize the state there is a potential to reveal abuses of power and, given the power imbalance, very little risk that undeserved consequences will befall those being studied.
• An EFF client, Project Recon, used an AI system to review massive volumes of transcripts of prison parole hearings to identify biased parole decisions. This innovative use of technology to identify systemic biases, including racial disparities, is the type of AI use we should support and encourage.

It is not a coincidence that the best examples of positive uses of AI come in places where experts, with access to infrastructure to help them use the technology and the requisite experience to evaluate the results, are involved. Moreover, academic researchers are already accustomed to explaining what they have done and being transparent about it—and it has been hard won knowledge that ethics are a vital step in work like this.

Nor is it a coincidence that other beneficial uses involve specific, discrete solutions to problems faced by those whose needs are often unmet by traditional channels or vendors. The ultimate outcome is beneficial, but it is moderated by human expertise and/or tailored to specific needs.

Context Matters

It can be very tempting—and easy—to make a blanket determination about something, especially when the stakes seem so high. But we urge everyone—users, policymakers, the companies themselves—to cut through the hype. In the meantime, EFF will continue to work against the harms caused by AI while also making sure that beneficial uses can advance.

Long before generative AI, copyright holders warned that new technologies for reading and analyzing information would destroy creativity. Internet search engines, they argued, were infringement machines—tools that copied copyrighted works at scale without permission. As they had with earlier information technologies like the photocopier and the VCR, copyright owners sued.

Courts disagreed. They recognized that copying works in order to understand, index, and locate information is a classic fair use—and a necessary condition for a free and open internet.

Today, the same argument is being recycled against AI. It’s whether copyright owners should be allowed to control how others analyze, reuse, and build on existing works.

Fair Use Protects Analysis—Even When It’s Automated

U.S. courts have long recognized that copying for purposes of analysis, indexing, and learning is a classic fair use. That principle didn’t originate with artificial intelligence. It doesn’t disappear just because the processes are performed by a machine.

Copying works in order to understand them, extract information from them, or make them searchable is transformative and lawful. That’s why search engines can index the web, libraries can make digital indexes, and researchers can analyze large collections of text and data without negotiating licenses from millions of rightsholders. These uses don’t substitute for the original works; they enable new forms of knowledge and expression.

Training AI models fits squarely within that tradition. An AI system learns by analyzing patterns across many works. The purpose of that copying is not to reproduce or replace the original texts, but to extract statistical relationships that allow the AI system to generate new outputs. That is the hallmark of a transformative use. 

Attacking AI training on copyright grounds misunderstands what’s at stake. If copyright law is expanded to require permission for analyzing or learning from existing works, the damage won’t be limited to generative AI tools. It could threaten long-standing practices in machine learning and text-and-data mining that underpin research in science, medicine, and technology. 

Researchers already rely on fair use to analyze massive datasets such as scientific literature. Requiring licenses for these uses would often be impractical or impossible, and it would advantage only the largest companies with the money to negotiate blanket deals. Fair use exists to prevent copyright from becoming a barrier to understanding the world. The law has protected learning before. It should continue to do so now, even when that learning is automated. 

A Road Forward For AI Training And Fair Use 

One court has already shown how these cases should be analyzed. In Bartz v. Anthropic, the court found that using copyrighted works to train an AI model is a highly transformative use. Training is a kind of studying how language works—not about reproducing or supplanting the original books. Any harm to the market for the original works was speculative. 

The court in Bartz rejected the idea that an AI model might infringe because, in some abstract sense, its output competes with existing works. While EFF disagrees with other parts of the decision, the court’s ruling on AI training and fair use offers a good approach. Courts should focus on whether training is transformative and non-substitutive, not on fear-based speculation about how a new tool could affect someone’s market share. 

AI Can Create Problems, But Expanding Copyright Is the Wrong Fix 

Workers’ concerns about automation and displacement are real and should not be ignored. But copyright is the wrong tool to address them. Managing economic transitions and protecting workers during turbulent times are core functions of government. Copyright law doesn’t help with those tasks in the slightest. Expanding copyright control over learning and analysis won’t stop new forms of worker automation—it never has. But it will distort copyright law and undermine free expression. 

Broad licensing mandates may also do harm by entrenching the current biggest incumbent companies. Only the largest tech firms can afford to negotiate massive licensing deals covering millions of works. Smaller developers, research teams, nonprofits, and open-source projects will all get locked out. Copyright expansion won’t restrain Big Tech—it will give it a new advantage.  

Fair Use Still Matters

Learning from prior work is foundational to free expression. Rightsholders cannot be allowed to control it. Courts have rejected that move before, and they should do so again.

Search, indexing, and analysis didn’t destroy creativity. Nor did the photocopier, nor the VCR. They expanded speech, access to knowledge, and participation in culture. Artificial intelligence raises hard new questions, but fair use remains the right starting point for thinking about training.

We're taking part in Copyright Week, a series of actions and discussions supporting key principles that should guide copyright policy. Every day this week, various groups are taking on different elements of copyright law and policy, and addressing what's at stake, and what we need to do to make sure that copyright promotes creativity and innovation.

In the Netflix/Spotify/Amazon era, many of us access copyrighted works purely in digital form – and that means we rarely have the chance to buy them. Instead, we are stuck renting them, subject to all kinds of terms and conditions. And because the content is digital, reselling it, lending it, even preserving it for your own use inevitably requires copying. Unfortunately, when it comes to copying digital media, US copyright law has pretty much lost the plot.

As we approach the 50th anniversary of the 1976 Copyrights, the last major overhaul of US copyright law, we’re not the only ones wondering if it’s time for the next one. It’s a high-risk proposition, given the wealth and influence of entrenched copyright interests who will not hesitate to send carefully selected celebrities to argue for changes that will send more money, into fewer pockets, for longer terms. But it’s equally clear that and nowhere is that more evident than the waning influence of Section 109, aka the first sale doctrine.

First sale—the principle that once you buy a copyrighted work you have the right to re-sell it, lend it, hide it under the bed, or set it on fire in protest—is deeply rooted in US copyright law. Indeed, in an era where so many judges are looking to the Framers for guidance on how to interpret current law, it’s worth noting that the first sale principles (also characterized as “copyright exhaustion”) can be found in the earliest copyright cases and applied across the rights in the so-called “copyright bundle.”

Unfortunately, courts have held that first sale, at least as it was codified in the Copyright Act, only applies to distribution, not reproduction. So even if you want to copy a rented digital textbook to a second device, and you go through the trouble of deleting it from the first device, the doctrine does not protect you.

We’re all worse off as a result. Our access to culture, from hit songs to obscure indie films, are mediated by the whims of major corporations. With physical media the first sale principle built bustling second hand markets, community swaps, and libraries—places where culture can be shared and celebrated, while making it more affordable for everyone.

And while these new subscription or rental services have an appealing upfront cost, it comes with a lot more precarity. If you love rewatching a show, you may be chasing it between services or find it is suddenly unavailable on any platform. Or, as fans of Mad Men or Buffy the Vampire Slayer know, you could be stuck with a terrible remaster as the only digital version available

Last year we saw one improvement with California Assembly Bill 2426 taking effect. In California companies must now at least disclose to potential customers if a “purchase” is a revocable license—i.e. If they can blow it up after you pay. A story driving this change was Ubisoft revoking access to “The Crew” and making customers’ copies unplayable a decade after launch. 

On the federal level, EFF, Public Knowledge, and 15 other public interest organizations backed Sen. Ron Wyden’s message to the FTC to similarly establish clear ground rules for digital ownership and sales of goods. Unfortunately FTC Chairman Andrew Ferguson has thus far turned down this easy win for consumers.

As for the courts, some scholars think they have just gotten it wrong. We agree, but it appears we need Congress to set them straight. The Copyright Act might not need a complete overhaul, but Section 109 certainly does. The current version hurts consumers, artists, and the millions of ordinary people who depend on software and digital works every day for entertainment, education, transportation, and, yes, to grow our food. 

We realize this might not be the most urgent problem Congress confronts in 2026—to be honest, we wish it was—but it’s a relatively easy one to solve. That solution could release a wave of new innovation, and equally importantly, restore some degree of agency to American consumers by making them owners again.

We're taking part in Copyright Week, a series of actions and discussions supporting key principles that should guide copyright policy. Every day this week, various groups are taking on different elements of copyright law and policy, and addressing what's at stake, and what we need to do to make sure that copyright promotes creativity and innovation.

Copyright owners increasingly claim more draconian copyright law and policy will fight back against big tech companies. In reality, copyright gives the most powerful companies even more control over creators and competitors. Today’s copyright policy concentrates power among a handful of corporate gatekeepers—at everyone else’s expense. We need a system that supports grassroots innovation and emerging creators by lowering barriers to entry—ultimately offering all of us a wider variety of choices.

Pro-monopoly regulation through copyright won’t provide any meaningful economic support for vulnerable artists and creators. Because of the imbalance in bargaining power between creators and publishing gatekeepers, trying to help creators by giving them new rights under copyright law is like trying to help a bullied kid by giving them more lunch money for the bully to take.

Entertainment companies’ historical practices bear out this concern. For example, in the late-2000’s to mid-2010’s, music publishers and recording companies struck multimillion-dollar direct licensing deals with music streaming companies and video sharing platforms. Google reportedly paid more than $400 million to a single music label, and Spotify gave the major record labels a combined 18 percent ownership interest in its now- $100 billion company. Yet music labels and publishers frequently fail to share these payments with artists, and artists rarely benefit from these equity arrangements. There’s no reason to think that these same companies would treat their artists more fairly now.

AI Training

In the AI era, copyright may seem like a good way to prevent big tech from profiting from AI at individual creators’ expense—it’s not. In fact, the opposite is true. Developing a large language model requires developers to train the model on millions of works. Requiring developers to license enough AI training data to build a large language model would  limit competition to all but the largest corporations—those that either have their own trove of training data or can afford to strike a deal with one that does. This would result in all the usual harms of limited competition, like higher costs, worse service, and heightened security risks. New, beneficial AI tools that allow people to express themselves or access information.

For giant tech companies that can afford to pay, pricey licensing deals offer a way to lock in their dominant positions in the generative AI market by creating prohibitive barriers to entry.

Legacy gatekeepers have already used copyright to stifle access to information and the creation of new tools for understanding it. Consider, for example, Thomson Reuters v. Ross Intelligence, the first of many copyright lawsuits over the use of works train AI. ROSS Intelligence was a legal research startup that built an AI-based tool to compete with ubiquitous legal research platforms like Lexis and Thomson Reuters’ Westlaw. ROSS trained its tool using “West headnotes” that Thomson Reuters adds to the legal decisions it publishes, paraphrasing the individual legal conclusions (what lawyers call “holdings”) that the headnotes identified. The tool didn’t output any of the headnotes, but Thomson Reuters sued ROSS anyways. A federal appeals court is still considering the key copyright issues in the case—which EFF weighed in on last year. EFF hopes that the appeals court will reject this overbroad interpretation of copyright law. But in the meantime, the case has already forced the startup out of business, eliminating a would-be competitor that might have helped increase access to the law.

Requiring developers to license AI training materials benefits tech monopolists as well. For giant tech companies that can afford to pay, pricey licensing deals offer a way to lock in their dominant positions in the generative AI market by creating prohibitive barriers to entry. The cost of licensing enough works to train an LLM would be prohibitively expensive for most would-be competitors.

The DMCA’s “Anti-Circumvention” Provision

The Digital Millennium Copyright Act’s “anti-circumvention” provision is another case in point. Congress ostensibly passed the DMCA to discourage would-be infringers from defeating Digital Rights Management (DRM) and other access controls and copy restrictions on creative works.

Section 1201 has been used to block competition and innovation in everything from printer cartridges to garage door openers

In practice, it’s done little to deter infringement—after all, large-scale infringement already invites massive legal penalties. Instead, Section 1201 has been used to block competition and innovation in everything from printer cartridges to garage door openers, videogame console accessories, and computer maintenance services. It’s been used to threaten hobbyists who wanted to make their devices and games work better. And the problem only gets worse as software shows up in more and more places, from phones to cars to refrigerators to farm equipment. If that software is locked up behind DRM, interoperating with it so you can offer add-on services may require circumvention. As a result, manufacturers get complete control over their products, long after they are purchased, and can even shut down secondary markets (as Lexmark did for printer ink, and Microsoft tried to do for Xbox memory cards.)

Giving rights holders a veto on new competition and innovation hurts consumers. Instead, we need balanced copyright policy that rewards consumers without impeding competition.

We're taking part in Copyright Week, a series of actions and discussions supporting key principles that should guide copyright policy. Every day this week, various groups are taking on different elements of copyright law and policy, and addressing what's at stake, and what we need to do to make sure that copyright promotes creativity and innovation.

There’s a crisis of creativity in mainstream American culture. We have fewer and fewer studios and record labels and fewer and fewer platforms online that serve independent artists and creators.  

At its core, copyright is a monopoly right on creative output and expression. It’s intended to allow people who make things to make a living through those things, to incentivize creativity. To square the circle that is “exclusive control over expression” and “free speech,” we have fair use.

However, we aren’t just seeing artists having a time-limited ability to make money off of their creations. We are also seeing large corporations turn into megacorporations and consolidating huge stores of copyrights under one umbrella. When the monopoly right granted by copyright is compounded by the speed and scale of media company mergers, we end up with a crisis in creativity. 

People have been complaining about the lack of originality in Hollywood for a long time. What is interesting is that the response from the major studios has rarely, especially recently, to invest in original programming. Instead, they have increased their copyright holdings through mergers and acquisitions. In today’s consolidated media world, copyright is doing the opposite of its intended purpose: instead of encouraging creativity, it’s discouraging it. The drive to snap up media franchises (or “intellectual properties”) that can generate sequels, reboots, spinoffs, and series for years to come has crowded out truly original and fresh creativity in many sectors. And since copyright terms last so long, there isn’t even a ticking clock to force these corporations to seek out new original creations. 

In theory, the internet should provide a counterweight to this problem by lowering barriers to entry for independent creators. But as online platforms for creativity likewise shrink in number and grow in scale, they have closed ranks with the major studios.  

It’s a betrayal of the promise of the internet: that it should be a level playing field where you get to decide what you want to do, watch, listen to, read. And our government should be ashamed for letting it happen.  

We're taking part in Copyright Week, a series of actions and discussions supporting key principles that should guide copyright policy. Every day this week, various groups are taking on different elements of copyright law and policy, and addressing what's at stake, and what we need to do to make sure that copyright promotes creativity and innovation.

Imagine every post online came with a bounty of up to $150,000 paid to anyone who finds it violates opaque government rules—all out of the pocket of the platform. Smaller sites could be snuffed out, and big platforms would avoid crippling liability by aggressively blocking, taking down, and penalizing speech that even possibly violates these rules. In turn, users would self-censor, and opportunists would turn accusations into a profitable business.

This dystopia isn’t a fantasy, it’s close to how U.S. copyright’s broken statutory damages regime actually works.

Copyright includes "statutory damages,” which means letting a jury decide how big of a penalty the defendant will have to pay—anywhere from $200 to $150,000 per work—without the jury necessarily seeing any evidence of actual financial losses or illicit profits. In fact, the law gives judges and juries almost no guidelines on how to set damages. This is a huge problem for online speech.

One way or another, everyone builds on the speech of others when expressing themselves online: quoting posts, reposting memes, sharing images from the news. For some users, re-use is central to their online expression: parodists, journalists, researchers, and artists use others’ words, sounds, and images as part of making something new every day. Both these users and the online platforms they rely on risk unpredictable, potentially devastating penalties if a copyright holder objects to some re-use and a court disagrees with the user’s well-intentioned efforts.

On Copyright Week, we like to talk about ways to improve copyright law. One of the most important would be to fix U.S. copyright’s broken statutory damages regime. In other areas of civil law, the courts have limited jury-awarded punitive damages so that they can’t be far higher than the amount of harm caused. Extremely large jury awards for fraud, for example, have been found to offend the Constitution’s Due Process Clause. But somehow, that’s not the case in copyright—some courts have ruled that Congress can set damages that are potentially hundreds of times greater than actual harm.

Massive, unpredictable damages awards for copyright infringement, such as a $222,000 penalty for sharing 24 music tracks online, are the fuel that drives overzealous or downright abusive takedowns of creative material from online platforms. Capricious and error-prone copyright enforcement bots, like YouTube’s Content ID, were created in part to avoid the threat of massive statutory damages against the platform. Those same damages create an ever-present bias in favor of major rightsholders and against innocent users in the platforms’ enforcement decisions. And they stop platforms from addressing the serious problems of careless and downright abusive copyright takedowns.

By turning litigation into a game of financial Russian roulette, statutory damages also discourage artistic and technological experimentation at the boundaries of fair use. None but the largest corporations can risk ruinous damages if a well-intentioned fair use crosses the fuzzy line into infringement.

“But wait”, you might say, “don’t legal protections like fair use and the safe harbors of the Digital Millennium Copyright Act protect users and platforms?” They do—but the threat of statutory damages makes that protection brittle. Fair use allows for many important re-uses of copyrighted works without permission. But fair use is heavily dependent on circumstances and can sometimes be difficult to predict when copyright is applied to new uses. Even well-intentioned and well-resourced users avoid experimenting at the boundaries of fair use when the cost of a court disagreeing is so high and unpredictable.

Many reforms are possible. Congress could limit statutory damages to a multiple of actual harm. That would bring U.S. copyright in line with other countries, and with other civil laws like patent and antitrust. Congress could also make statutory damages unavailable in cases where the defendant has a good-faith claim of fair use, which would encourage creative experimentation. Fixing fair use would make many of the other problems in copyright law more easily solvable, and create a fairer system for creators and users alike.

This year, we fought back against the return of a terrible idea that hasn’t improved with age: site blocking laws. 

More than a decade ago, Congress tried to pass SOPA and PIPA—two sweeping bills that would have allowed the government and copyright holders to quickly shut down entire websites based on allegations of piracy. The backlash was massive. Internet users, free speech advocates, and tech companies flooded lawmakers with protests, culminating in an “Internet Blackout” on January 18, 2012. Turns out, Americans don’t like government-run internet blacklists. The bills were ultimately shelved.  

But we’ve never believed they were gone for good. The major media and entertainment companies that backed site blocking in the US in 2012 turned to pushing for site-blocking laws in other countries. Rightsholders continued to ask US courts for site-blocking orders, often winning them without a new law. And sure enough, the Motion Picture Association (MPA) and its allies have asked Congress to try again. 

There were no less than three Congressional drafts of site-blocking legislation. Representative Zoe Lofgren kicked off the year with the Foreign Anti-Digital Piracy Act (FADPA). Fellow House of Representatives member Darrell Issa also claimed to be working on a bill that would make it offensively easy for a studio to block your access to a website based solely on the belief that there is infringement happening. Not to be left out, the Senate Judiciary Committee produced the terribly named Block BEARD Act.  

None of these three attempts to fundamentally alter the way you experience the internet moved too far after their press releases. But the number tells us that there is, once again, an appetite among major media conglomerates and politicians to resurrect SOPA/PIPA from the dead.  

None of these proposals fixes the flaws of SOPA/PIPA, and none ever could. Site blocking is a flawed idea and a disaster for free expression that no amount of rewriting will fix. There is no way to create a fast lane for removing your access to a website that is not a major threat to the open web. Just as we opposed SOPA/PIPA over ten years ago, we oppose these efforts.  

Site blocking bills seek to build a new infrastructure of censorship into the heart of the internet. They would enable court orders directed to the organizations that make the internet work, like internet service providers, domain name resolvers, and reverse proxy services, compelling them to help block US internet users from visiting websites accused of copyright infringement. The technical means haven’t changed much since 2012. - tThey involve blocking Internet Protocol addresses or domain names of websites. These methods are blunt—sledgehammers rather than scalpels. Today, many websites are hosted on cloud infrastructure or use shared IP addresses. Blocking one target can mean blocking thousands of unrelated sites. That kind of digital collateral damage has already happened in Austria, Italy, South Korea, France, and in the US, to name just a few.  

Given this downside, one would think the benefits of copyright enforcement from these bills ought to be significant. But site blocking is trivially easy to evade. Determined site owners can create the same content on a new domain within hours. Users who want to see blocked content can fire up a VPN or change a single DNS setting to get back online.  

The limits that lawmakers have proposed to put on these laws are an illusion. While ostensibly aimed at “foreign” websites, they sweep in any website that doesn’t conspicuously display a US origin, putting anonymity at risk. And despite the rhetoric of MPA and others that new laws would be used only by responsible companies against the largest criminal syndicates, laws don’t work that way. Massive new censorship powers invite abuse by opportunists large and small, and the costs to the economy, security, and free expression are widely borne. 

It’s time for Big Media and its friends in Congress to drop this flawed idea. But as long as they keep bringing it up, we’ll keep on rallying internet users of all stripes to fight it. 

This article is part of our Year in Review series. Read other articles about the fight for digital rights in 2025.

State legislatures—from Olympia, WA, to Honolulu, HI, to Tallahassee, FL, and everywhere in between—kept EFF’s state legislative team busy throughout 2025.

We saw some great wins and steps forward this year. Washington became the eighth state to enshrine the right to repair. Several states stepped up to protect the privacy of location data, with bills recognizing your location data isn't just a pin on a map—it's a powerful tool that reveals far more than most people realize. Other state legislators moved to protect health privacy. And California passed a law making it easier for people to exercise their privacy rights under the state’s consumer data privacy law.

Several states also took up debates around how to legislate and regulate artificial intelligence and its many applications. We’ll continue to work with allies in states including California and Colorado to proposals that address the real harms from some uses of AI, without infringing on the rights of creators and individual users.

We’ve also fought some troubling bills in states across the country this year. In April, Florida introduced a bill that would have created a backdoor for law enforcement to have easy access to messages if minors use encrypted platforms. Thankfully, the Florida legislature did not pass the bill this year. But it should set off serious alarm bells for anyone who cares about digital rights. And it was just one of a growing set of bills from states that, even when well-intentioned, threaten to take a wrecking ball to privacy, expression, and security in the name of protecting young people online.

Take, for example, the burgeoning number of age verification, age gating, age assurance, and age estimation bills. Instead of making the internet safer for children, these laws can incentivize or intersect with existing systems that collect vast amounts of data to force all users—regardless of age—to verify their identity just to access basic content or products. South Dakota and Wyoming, for example, are requiring any website that hosts any sexual content to implement age verification measures. But, given the way those laws are written, that definition could include essentially any site that allows user-generated or published content without age-based gatekeeping access. That could include everyday resources such as social media networks, online retailers, and streaming platforms.

Lawmakers, not satisfied with putting age gates on the internet, are also increasingly going after VPNs (virtual private networks) to prevent anyone from circumventing these new digital walls. VPNs are not foolproof tools—and they shouldn’t be necessary to access legally protected speech—but they should be available to people who want to use them. We will continue to stand against these types of bills, not just for the sake of free expression, but to protect the free flow of information essential to a free society.

This article is part of our Year in Review series. Read other articles about the fight for digital rights in 2025.

A functioning patent system depends on one basic principle: bad patents must be challengeable. In 2025, that principle was repeatedly tested—by Congress, by the U.S. Patent and Trademark Office (USPTO), and by a small number of large patent owners determined to weaken public challenges. 

Two damaging bills, PERA and PREVAIL, were reintroduced in Congress. At the same time, USPTO attempted a sweeping rollback of inter partes review (IPR), one of the most important mechanisms for challenging wrongly granted patents. 

EFF pushed back—on Capitol Hill, inside the Patent Office, and alongside thousands of supporters who made their voices impossible to ignore.

Congress Weighed Bills That Would Undo Core Safeguards

The Patent Eligibility Restoration Act, or PERA, would overturn the Supreme Court’s Alice and Myriad decisions—reviving patents on abstract software ideas, and even allowing patents on isolated human genes. PREVAIL, introduced by the same main sponsors in Congress, would seriously weaken the IPR process by raising the burden of proof, limiting who can file challenges, forcing petitioners to surrender court defenses, and giving patent owners new ways to rewrite their claims mid-review.

Together, these bills would have dismantled much of the progress made over the last decade. 

We reminded Congress that abstract software patents—like those we’ve seen on online photo contests, upselling prompts, matchmaking, and scavenger hunts—are exactly the kind of junk claims patent trolls use to threaten creators and small developers. We also pointed out that if PREVAIL had been law in 2013, EFF could not have brought the IPR that crushed the so-called “podcasting patent.” 

EFF’s supporters amplified our message, sending thousands of messages to Congress urging lawmakers to reject these bills. The result: neither bill advanced to the full committee. The effort to rewrite patent law behind closed doors stalled out once public debate caught up with it. 

Patent Office Shifts To An “Era of No”

Congress’ push from the outside was stymied, at least for now. Unfortunately, what may prove far more effective is the push from within by new USPTO leadership, which is working to dismantle systems and safeguards that protect the public from the worst patents.

Early in the year, the Patent Office signaled it would once again lean more heavily on procedural denials, reviving an approach that allowed patent challenges to be thrown out basically whenever there was an ongoing court case involving the same patent. But the most consequential move came later: a sweeping proposal unveiled in October that would make IPR nearly unusable for those who need it most.

2025 also marked a sharp practical shift inside the agency. Newly appointed USPTO Director John Squires took personal control of IPR institution decisions, and rejected all 34 of the first IPR petitions that came across his desk. As one leading patent blog put it, an “era of no” has been ushered in at the Patent Office. 

The October Rulemaking: Making Bad Patents Untouchable

The USPTO’s proposed rule changes would: 

• Force defendants to surrender their court defenses if they use IPR—an intense burden for anyone actually facing a lawsuit. 
• Make patents effectively unchallengeable after a single prior dispute, even if that challenge was limited, incomplete, or years out of date.
• Block IPR entirely if a district court case is projected to move faster than the Patent Trial and Appeal Board (PTAB). 

These changes wouldn’t “balance” the system as USPTO claims—they would make bad patents effectively untouchable. Patent trolls and aggressive licensors would be insulated, while the public would face higher costs and fewer options to fight back. 

We sounded the alarm on these proposed rules and asked supporters to register their opposition. More than 4,000 of you did—thank you! Overall, more than 11,000 comments were submitted. An analysis of the comments shows that stakeholders and the public overwhelmingly oppose the proposal, with 97% of comments weighing in against it. 

In those comments, small business owners described being hit with vague patents they could never afford to fight in court. Developers and open-source contributors explained that IPR is often the only realistic check on bad software patents. Leading academics, patient-advocacy groups, and major tech-community institutions echoed the same point: you cannot issue hundreds of thousands of patents a year and then block one of the only mechanisms that corrects the mistakes.

The Linux Foundation warned that the rules “would effectively remove IPRs as a viable mechanism” for developers.

GitHub emphasized the increased risk and litigation cost for open-source communities.

Twenty-two patent law professors called the proposal unlawful and harmful to innovation.

Patients for Affordable Drugs detailed the real-world impact of striking invalid pharmaceutical patents, showing that drug prices can plummet once junk patents are removed.

Heading Into 2026

The USPTO now faces thousands of substantive comments. Whether the agency backs off or tries to push ahead, EFF will stay engaged. Congress may also revisit PERA, PREVAIL, or similar proposals next year. Some patent owners will continue to push for rules that shield low-quality patents from any meaningful review.

But 2025 proved something important: When people understand how patent abuse affects developers, small businesses, patients, and creators, they show up—and when they do, their actions can shape what happens next. 

This article is part of our Year in Review series. Read other articles about the fight for digital rights in 2025.