Courts have for two decades found that linking and embedding someone else’s web content, be it a photo, music, or an article, doesn’t violate copyright law–the entity that controls the server that hosts a copyrighted work, not the user or website that merely directs others to it, is directly liable if the content turns out to be infringing.

News publisher Emmerich Newspapers sought to convince the Fifth Circuit Court of Appeals to chart a new and dangerous course, arguing that an aggregator website that published links to its copyrighted articles was in effect “displaying” them and can be directly liable for infringement. EFF, along with several other public interest organizations and trade associations, filed a brief urging the court to follow multiple other circuits and reject that theory.

Fortunately, the Fifth Circuit Court of Appeals did just that. While it rejected the server test–the rule courts have used to determine copyright liability rests with whoever serves up the content–the court came to the same practical conclusion by focusing on who is responsible for transmitting content. 

Applying that test, the court found that pointing or directing a user’s browser to request and receive the copyright owner’s own copy residing on its computers does not involve transmitting or communicating the content. “Although we take different routes to get there, both the server test and the test we announce end up in a similar place: a website cannot transmit a work that it does not have,” the court said. 

We told the court that accepting Emmerich's theory would make the common act of embedding links a legally fraught activity, one that many websites might be unwilling to risk, which would seriously damage the internet as a tool for creating and disseminating ideas and knowledge,

We applaud the court’s decision–even though it applied a different test, it correctly concluded that a user linking pictures, video, or articles isn’t in charge of transmitting that content to the world. The user doesn’t control what’s located on the other end of the link—that’s up to the person who controls the server.

Emmerich also claimed linking violates the Digital Millennium Copyright Act (DMCA), arguing its URLs were copyright management information (CMI) and when the aggregator displayed Emmerich’s articles under its own URL, it tampered with Emmerich’s CMI, which violates the DMCA. 

Under that logic, unsuspecting internet users could face ruinous legal risk for doing something as simple as using a link shortener, particularly given potential statutory penalties of up to $25,000 per violation.  

In our brief, we told the court that URLs don’t necessarily equate to a copyrighted work or provide sufficient information about the nature of the underlying content, making it highly unlikely that anyone would expect a URL to contain CMI. Quoting EFF’s brief, the court concluded that URLs are first and foremost a locational reference tool and while it may be possible for a URL to contain CMI, the bar to that conclusion is high.

Overall, this was a good and sensible decision that will protect ordinary online expression, communication, and access to knowledge. Hopefully this issue is laid to rest at last.

The history of technology is rife with copyright panics.  In the 1980s, major rightsholders ran to Congress and the courts, claiming that videotape recorders (VTR) were “to the American film producer and the American public as the Boston strangler is to the woman home alone.” Then, the Supreme Court declined to embrace the hype, noting that the VTR was capable of all kinds of non-infringing uses, like time-shifting and cautioning courts to avoid rewriting copyright law in response to new technologies. We believe that courts now should be similarly wary about the hype surrounding AI.

DONATE TO EFF 

ensure ai serves people, not power

Hollywood’s hyperbole has echoed that of composer John Phillip Sousa, who claimed in 1906 that the player piano and the gramophone would destroy music composition; portrait artists who feared the camera would replace the paintbrush. None of these things happened. Cameras, for example, sparked a resurgence of portraiture and, by making it possible for more people to create images, led to unexpected developments—like the rise of photojournalism.

New markets, new ideas, and new creators are actually what copyright is supposed to promote, not restrict. Using copyright to lock in existing gatekeepers and massive rightsholders’ profits helps neither the public nor individual artists.

Generative AI has sparked the latest wave of anxiety and with it a massive wave of litigation. In multiple cases around the U.S. and the world, rightsholders are asking courts to do precisely what the Supreme Court warned against: dramatically expand copyright protections based in substantial part on hyperbole and speculation. They should decline to do so.

Copyright owners claim that unless courts abandon 300-year-old copyright principles—and give rightsholders the power to control non-infringing works created by others—an imagined flood of AI-generated works will devastate creative markets. Under this “market dilution” theory, building generative AI tools cannot be fair use because those tools might be encourage the proliferation of competing works.

As EFF has explained to the courts in multiple amicus briefs in Concord Music Group, Inc. v. Anthropic PBC and In re Mosaic LLM Litigation, that’s not how copyright works. In fact, accepting this theory would undermine copyright’s constitutional purpose: promoting the creation of expressive works for the public’s benefit. Because copyright law is designed to encourage others to build freely on existing works, it punishes infringement, not competition. The “market dilution” theory would eviscerate not only the fair use doctrine, but also other limits on copyright that work specifically to prevent rightsholders from unfairly suppressing competition by claiming broad ownership over tropes, genres, styles, and so on. In other words, publishers would wield unchecked veto power over any expression that might conceivably compete with a work they own.

The result? Art doesn’t get created, ideas are never expressed, and we’re all worse off. Copyright shouldn’t be a tool to silence future creative competitors—whether or not they use AI in their work.

And the plaintiffs in these cases get at least two other things wrong. First, research shows that large generative AI models are unlikely to produce infringing works because the more data on which a model is trained, the less any individual training example matters to any particular output.

Second, AI tools aren’t necessarily displacing human creativity. To take a just a few examples:

• Boston-based artist Nettrice Gaskins uses AI to create Afro-futurist art, including a portrait of Octavia Butler displayed at the San Francisco Airport

• Indian artists Prateek Arora and Varun Gupta use generative AI to reimagine Western science fiction.

• Philadelphia-based artist Alex Smith uses generative AI to reimagine Afrofuturism with queer, plus-sized Black superheroes.

• Ana Miljački, a professor of architecture at MIT, used generative AI to create a “non-liner documentary” film on Yugoslav World War II memorials and the values they embodied.

• A research-creation project used AI generated visual art to both amplify the voices of activists in the Iran Woman Life Freedom Movement and evaluate AI’s role in sociopolitical advocacy through art.

• AI company Bronze works with musicians like Disclosure and Jai Paul to create songs that never sound the same when played back twice, challenging audience conceptions of what music could be.

It is not the place of courts to say these people are not artists or that AI cannot augment human creativity in a positive way.

Given this range of experimentation, courts should be reluctant to decide in advance what tools do and do not foster “human creativity.” Like the VTR, large language models are general purpose tools, used by humans to do a broad variety of things far beyond generating lyrics. The effects of this particular technological innovation will doubtless be far-reaching, disruptive, and potentially harmful for some—but distorting copyright law is not the way to address those harms.

This post is part 2 in a series about automated content moderation. Read the first post here.

When whistleblower Frances Haugen leaked a set of documents from Meta in 2020, among the revelations was a jarring statistic: The company’s algorithms designed to detect terrorist content incorrectly deleted nonviolent Arabic-language content 77 percent of the time, while failing to detect hate speech under the company’s own policies in many instances. Meta’s own transparency report released later that year demonstrated similar findings. Five years later, researchers in the region report that overzealous moderation remains a problem, while paths to remedy have all but collapsed.

Where these systems are faltering in Arabic, they’re positively failing in less-resourced languages. As a 2025 report from the Center for Democracy and Technology found, labeled datasets in certain languages and dialects such as Maghrebi Arabic and Kiswahili contain inconsistencies, bias, and inaccuracies due to the limited hiring of annotators who actually speak the languages as well as shifts in the languages themselves. An investigation into ChatGPT’s outputs in several low-resource languages demonstrates the depth of problem.

But language disparities are just one of several concerns as automated moderation becomes more widespread. From the systemic suppression of content from Palestine to the repeated misclassification of LGBTQ+ content as adult or explicit material, these varied examples demonstrate the risks of overreliance on automated moderation—and the need for stronger safeguards.

Transparency, Cultural Competence, Appeals

As we discussed in Part 1 of this series, automated systems can process content at a scale that humans never could, potentially enabling better moderation at scale and alleviating the psychological load on ill-paid moderators whose jobs require them to view incredibly disturbing content. But automated systems also reproduce existing biases, struggle to understand context, and often make mistakes that disproportionately affect journalists, activists, artists, and other vulnerable and marginalized communities.

As Rachel Griffin wrote in 2023, “Perfectly accurate moderation is not only technically out of reach but intrinsically impossible.” Despite those intrinsic flaws, there is a great deal companies, policymakers, and civil society can do to help ensure that highly-automated systems operate in ways that respect human rights, minimize predictable harms, and provide meaningful accountability when they fail. If companies are going to continue relying on automation to moderate users’ speech—and there is little reason to believe they won’t—then accountability must evolve alongside these technologies.

That evolution can start with committing to the Santa Clara Principles 2.0. These principles, first outlined in 2020 and re-launched in 2021 after substantial international input, reflect the needs and expectations of the global community and specifically address automation. The first Foundational Principle states:

Companies should ensure that human rights and due process considerations are integrated at all stages of the content moderation process, and should publish information outlining how this integration is made. Companies should only use automated processes to identify or remove content or suspend accounts, whether supplemented by human review or not, when there is sufficiently high confidence in the quality and accuracy of those processes. Companies should also provide users with clear and accessible methods of obtaining support in the event of content and account action. 

Drawing on the Santa Clara Principles 2.0, international human rights standards, and years of research documenting the shortcomings of automated moderation, we propose eight recommendations for policymakers thinking about regulation and companies deploying AI-assisted content moderation systems.

1. Automated technologies should help, not replace, human moderators. For example, automated systems can help flag and prioritize content for review, while humans can interpret context, handle sensitive cases, and refine system performance.
2. Companies must be transparent about when and how automation is used in content decisions.
3. Companies must regularly audit their automated systems for bias, with particular attention to low-resource languages, vulnerable and marginalized communities, and conflict zones.
4. Users must have the ability to appeal, and to provide context when they believe human or automated moderation decisions have wrongfully removed their content. Appeals should be promptly evaluated and decided by human moderators.
5. Companies should regularly assess the human rights impact of their moderation decisions, and issue public statements of the results
6. If they rely on third-party vendors, companies should carefully (and regularly) audit those vendors for compliance with these same principles
7. Lawmakers should avoid promoting and passing legislation that effectively or explicitly mandates automated moderation systems
8. Policymakers should also refrain from attempting to dictate platforms technical and design choices to favor or disfavor particular expression.

These recommendations understand that automated content moderation isn’t just a technical problem for clever engineers and product teams to solve. Because content moderation shapes public discourse and fundamental rights, its design and oversight must respond to the concerns of policymakers, civil society, independent researchers, and the communities most affected by these systems.

This is the second post in a 2-part series on automated content moderation. Read the first post here.

This blog post is part 1 of a 2-part series. The second part sets out recommendations for companies and policymakers.

Six years ago—one month into a global pandemic—we argued that the automated moderation processes many platforms were rapidly adopting should be highly transparent, easily appealable, and temporary. We warned that "protocols adopted in times of crisis often persist when the crisis is over."

That warning proved prescient. The use of automation and artificial intelligence (AI) to identify, flag, and moderate content has become the new norm—a permanent feature of how platforms govern speech online. In this two part series, we’re take stock of this new norm, and considering what platforms can and should do to ensure that AI serves online expression rather than stifling it.

A brief history of automated content moderation

From spam filtering and keyword blacklists to the hash-matching technologies used to identify child sexual abuse material and terrorist content, automated technologies have been used in commercial content moderation for many years. While these tools have long posed risks to freedom of expression, their use was, for quite some time, relatively limited in scope.

Then, in 2017, a blog post published by Facebook (now Meta) described the company's "fairly recent" use of artificial intelligence to identify, classify, and remove violent extremist content. At the same time, Facebook emphasized caution, noting that it did not want to suggest there was "any easy technical fix."

Just one year later, Mark Zuckerberg appeared before the U.S. Senate's Commerce and Judiciary Committees and disclosed that "99 percent of the ISIS and Al Qaida content" removed by Facebook was flagged by AI "before any human sees it." He also stated that Facebook was "developing A.I. tools that can identify certain classes of bad activity proactively and flag it for our team at Facebook." At the time, we raised concerns about the ethical implications of using AI in this manner.

Then came 2020. The sudden reduction of the human moderation workforce, combined with a dramatic increase in social media use—and with it, a surge in misinformation—created the perfect conditions for platforms to expand their reliance on AI-driven moderation. It quickly became apparent that companies'—and particularly Meta's—approach to moderation during the pandemic represented a backslide in transparency, freedom of expression, and access to remedy. The increased reliance on automation was a significant factor.

The costs and benefits of AI content moderation

We knew in 2020 that the use of AI to moderate content would present problems for online freedom of expression. Today, those problems are well-documented. A 2025 joint declaration by special rapporteurs and representatives of the United Nations (UN), Organization for Security and Co-operation in Europe (OSCE), Organization of American States (OAS), and African Commission on Human and Peoples’ Rights (ACHPR) states:

“The use of AI content moderation can lead to over-removal, discrimination and censorship. Reliance on inherently biased datasets and opaque training processes can amplify pre-existing inequalities, risking homogenisation of expression, and erasure of linguistic and cultural diversity.”

EFF and many of our allies have documented these impacts. For example, our 2019 paper co-authored with Witness and Syrian Archive examined the impact of extremist content regulations—and their implementation through automation and AI—on human rights documentation. A 2020 report from Human Rights Watch highlighted the consequences of these removals, noting: "There is no way of knowing how much potential evidence of serious crimes is disappearing without anyone's knowledge."

The Center for Democracy and Technology's recent series on content moderation in the Global South demonstrates persistent inequities in content moderation of four “low-resource” languages—so-called because the relative scarcity of training data makes it more difficult to develop equitable and accurate AI models for them. 

Content moderation often disproportionately impacts vulnerable and historically marginalized groups, and AI content moderation is no different. GLAAD recognizes the role AI plays in scaling content moderation but notes that “when moderation systems lack nuance, transparency, and human oversight, they can fail to curb harassment and wrongly suppress legitimate LGBTQ content.”

These failures are not incidental. They are a predictable consequence of deploying automated systems to make complex judgments about language, culture, context, and identity at scale.

All of that said, automated content moderation can offer important benefits. The primary one: helping to spare human content moderators who must review content that varies from whimsical to horrific, often for little pay and with devastating mental health consequences. Outsourcing this work to the bots can offer some relief—though it’s worth noting that the humans hired to train the AI models face a similar dynamic.

In addition, AI models could potentially be trained over time to be more precise, accurate, and dynamic, helping to mitigate over-censorship and disinformation. The jury is still out on whether this potential will be realized; what we do know is that new approaches to the persistent problem of over and under-enforcement are desperately needed.

Automated moderation is no longer an experiment

Getting the balance between real costs and potential benefits depends a lot on the details: how automated systems are designed, trained, implemented, and audited.  

Despite advances in the sophistication and scale of automated moderation systems, many of the transparency, accountability, and due process safeguards advocated by civil society, researchers, and human rights experts have yet to be fully realized. At the same time, automated systems have become increasingly central to how platforms enforce their rules and govern online speech.

The question today is not whether companies will use AI to moderate content, but under what conditions they should do so. And now as ever, the answer is not that the public should just trust that platforms’ deployment of increasingly powerful systems will serve, rather than inhibit online expression. In fact, as automated systems become more sophisticated and more deeply embedded in platform governance, the need for transparency and accountability becomes more urgent. 

This is part 1 of a 2-part series. You can read the second part here.

The Trump administration’s approach to AI safety, particularly the generative AI models that regularly grab headlines, has been haphazard at best. At worst, it’s unconstitutional. As EFF and our allies explained in an amicus brief, the Pentagon’s actions against one company, Anthropic, violate the First Amendment because they were motivated by the administration’s desire to punish an uncooperative company, not legitimate concerns about national security.

By and large, the Trump administration’s AI strategy has minimized regulation in the name of “winning” the global “race” to develop leading frontier models. It has pared back regulations intended to address even the most serious AI threats—like AI-enabled cyberattacks on government systems—to protect AI innovation.

Yet it has repeatedly singled out one AI company for arbitrary, heavy-handed rules and sanctions. For years, the federal government relied on Anthropic’s models for use in its classified systems. But after Anthropic resisted the government’s demands to use Anthropic’s models to autonomously kill people or spy on Americans, the government declared war on the “woke” company. It designated the company a “supply chain risk,” effectively banning agencies and government contractors from doing business with the company.

A court issued a preliminary injunction preventing these sanctions from taking effect, as EFF and other civil liberties organizations urged it to do in an amicus brief filed earlier this year. But absent judicial action, these sanctions would’ve cost the company hundreds of millions of dollars. Either way, it sent a clear signal that companies must adhere to the government’s wishes or face similar consequences.

As we explained in our brief filed today, these sanctions were clear retaliation for the company’s public refusal to allow the Pentagon to use its models to develop fully autonomous weapons and spy on Americans. This kind of retaliation is unconstitutional.

In a recent executive order, the Trump administration took its war on Anthropic even further, by imposing “export controls” that ban any foreign nationals from using Anthropic’s new Mythos and Fable models. To comply with this order, Anthropic shut down the models altogether.

These extreme measures were purportedly justified by security concerns. The administration said it feared that Anthropic’s Mythos-class models could be used to find and exploit existing vulnerabilities in software code—hardly a new feat for an LLM. Anthropic itself has contributed to public anxieties about its Mythos-class models, initially claiming that Mythos was too dangerous for public release and restricting access to a handful of partners. The company’s CEO called for a pause on AI development, citing fears that the technology was becoming too powerful.

But regulators should be cutting through the hype, not feeding it. Even if Mythos’s capabilities were a modest improvement over existing technology, others are already closing the gap. In other words, nothing about Mythos is so uniquely dangerous that it warrants exceptional export controls to protect the public. Yet other LLMs with similar offensive cybersecurity capabilities are not subject to export controls. Instead, the government has embraced a voluntary system in which companies are encouraged to submit models to the government for cybersecurity testing 30 days before releasing them to the public.

AI policy should be reasonably responsive to real-world risk, grounded in the realities of the technology, and no more burdensome than necessary to protect the public. But the government’s haphazard decision to impose export controls on Mythos-class models, while subjecting other AI models to nothing more than a voluntary, light-touch framework, meets none of these criteria. As leading cybersecurity experts and executives recently explained in an open letter, these sanctions prevent developers and security teams from using the best models to find and fix vulnerabilities before adversaries, armed with nearly as capable AI, can exploit them.

Decades Later, Code Is Still Speech

More importantly, export controls on important software tools like LLMs can undermine the free flow of digital communications and technologies that activists, innovators, and ordinary users desperately need. Freedom of expression requires access to these tools. Depriving the public of the best AI threatens our rights without making us any safer.

EFF has long opposed government efforts to restrict the publication of non-classified software to the general public. In the 1990s, EFF challenged export controls on encryption software, helping establish the principle that “code is speech,” protected by the First Amendment. Courts recognized that software is not just a functional tool—it’s a means of ideas, knowledge, and technical know-how. And they recognized that the government was overreaching in trying to restrict private developers from sharing their improvements in computer security with the public.

While AI models raise new questions, efforts to restrict access to them implicate the same constitutional and speech concerns as older efforts to restrict encryption. Export controls are uniquely susceptible to abuse. And they are especially suspect when they are unilaterally imposed without clear and fair standards.

Whether these export controls were another attempt to punish Anthropic or simply a misguided security measure, the public loses. The real cybersecurity risks of advanced AI may ultimately justify limited regulations to protect the public from legitimate threats. But whether the government ultimately chooses to heavily regulate the technology or hold off to promote innovation, its rules must be rational and evenhanded. 

• Read EFF's Amicus Brief in Anthropic v. Department of War