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.

California lawmakers are fast-tracking A.B. 1709—a sweeping bill that would ban anyone under 16 from using social media and force every user, regardless of age, to submit sensitive personal information before accessing social platforms.

That means that under A.B. 1709, social media companies will have to enact age gates to prohibit minors from accessing their services. The services may decide that complying with this bill means that Californians have to submit highly sensitive government-issued ID or biometric information to prove they are adults. In the name of “safety,” this bill would destroy online anonymity, expose sensitive personal data to breach and abuse, and replace parental decision-making with state-mandated censorship.

A.B. 1709 has already passed out of the Assembly Privacy and Judiciary Committees with nearly unanimous support. Its next stop is the Assembly Appropriations Committee, followed by a floor vote—likely within the next week.

Take action

Tell Your Representative to OPPOSE A.B. 1709

California Is About to Set a Dangerous Precedent for Online Censorship

By banning access to social media platforms for young people under 16, California is emulating Australia, where early results show exactly what EFF and other critics predicted: overblocking by platforms, leaving youth without support and even adults barred from access; major spikes in VPN use and other workarounds ranging from clever to desperate; and smaller platforms shutting down rather than attempting costly compliance with these sweeping bills.

California should not be racing to replicate those failures. After all, when California leads—especially on tech—other states follow. There is no reason for California to lead the nation into an unconstitutional social media ban that destroys privacy and harms youth.

Take action

Tell Your Representative to OPPOSE A.B. 1709

What’s Wrong With A.B. 1709?

Just about everything.

A.B. 1709 weaponizes legitimate parental concerns by using them to hand over even more censorship and surveillance power to the government. Beneath its shiny “protect the children” rhetoric, this bill is misguided, unconstitutional, and deeply harmful to users of all ages.

A.B. 1709 Recklessly Violates Free Speech Rights

The First Amendment protects the right to speak and access information, regardless of age. But by imposing a blanket ban on social media access, A.B. 1709 would cut off lawful speech for millions of California teenagers, while also forcing all users (adults and kids alike) to verify their ages before speaking or accessing information on social media. This will immensely and unconstitutionally chill Californians’ exercise of their First Amendment.

These mandates ignore longstanding Supreme Court precedent that protects young people’s speech and consistently find these bans unconstitutional. Banning young people entirely from social media is an extreme measure that doesn’t match the actual risks of online engagement. California simply does not have a valid interest in overriding parents’ and young people’s rights to decide for themselves how to use social media.

After all, age-verification technology is far from perfect. A.B. 1709’s reliance on imperfect age-verification technology will disproportionately silence marginalized communities—those whose IDs don’t match their presentation, those with disabilities, trans and gender non-conforming folks, and people of color—who are most likely to be wrongfully denied access by discriminatory systems.  

Finally, many people will simply refuse to give up their anonymity in order to access social media. Our right to anonymity has been a cornerstone of free expression since the founding of this country, and a pillar of online safety since the dawn of the internet. This is for good reason: it allows creativity, innovation, and political thought to flourish, and is essential for those who risk retaliation for their speech or associations. A.B. 1709 threatens to destroy it.

AB 1709 Needlessly Jeopardizes Everyone’s Privacy

A.B. 1709’s age-gating mandate also creates massive security risks by incentivizing platforms to force all of their users to hand over immutable biometric data and government IDs to third-party vendors. By creating centralized "honeypots" of sensitive information, the bill invites identity theft and permanent surveillance rather than actual safety. If we don’t trust tech companies with our private information now, we shouldn't pass a law that mandates we give them even more of it. 

We’ve already seen repeated data breaches involving age- and identity-verification services. Yet A.B. 1709 would require millions more Californians—including the youth this bill claims to protect—to feed their most sensitive data into this growing surveillance ecosystem. 

This is not the answer to online safety.

Take action

Tell Your Representative to OPPOSE A.B. 1709

AB 1709 Harms the Youth It Claims to Protect

While framed as a safety measure, this bill serves as a blunt instrument of censorship, severing vital lifelines for California’s young people. Besides being unconstitutional, banning young people from the internet is bad public policy. After all, social media sites are not just sources of entertainment; they provide crucial spaces for young people to explore their identities—whether by creating and sharing art, practicing religion, building community, or engaging in civic life. 

Social science indicates that moderate internet use is a net positive for teens’ development, and negative outcomes are usually due to either lack of access or excessive use. Social media provides essential spaces for civic engagement, identity exploration, and community building—particularly for LGBTQ+ and marginalized youth who may lack support in their physical environments. By replacing access to political news and health resources with state-mandated isolation, A.B. 1709 ignores the calls of young people themselves who favor digital literacy and education over restrictive government control.

Young people have been loud and clear that what they want is access and education—not censorship and control. They even drafted their own digital literacy education bill, A.B. 2071, which is currently before the California legislature! Instead of cutting off vital lifelines, we should support education measures that would arm them (and the adults in their lives) with the knowledge they need to explore online spaces safely.

AB 1709 Is Misguided and Won’t Work

In case you needed more reasons to oppose this bill.

• A.B. 1709 Replaces Parenting With Government Control. Families know there is no one-size-fits-all solution to parenting. But AB 1709 imposes one anyway, overriding parental decision-making with a blanket censorship prohibition. Parents who want to actively guide their children’s online experiences should be empowered, not relegated to the sidelines by a blunt state mandate.
• A.B. 1709 Strengthens Big Tech Instead of Challenging It. Supporters claim that this bill will rein in the major tech companies, but in fact, steep fines and costly compliance regimes disproportionately harm smaller platforms. Where large corporations can afford to absorb legal risk and shell out for expensive verification systems, smaller forums and emerging platforms cannot. We’ve already seen platforms shut down or geoblock entire states in response to age-gating laws. And when the small platforms shutter, where do all of those users—and their valuable data—go? Straight back to the biggest companies.
• A.B. 1709 Creates Expensive and Shady Bureaucracy During a Budget Crisis. California is facing a massive deficit, but A.B. 1709 would waste taxpayer dollars to fund a shadowy new "e-Safety Advisory Commission" to enforce this ban and dream up new ways to censor the internet. In addition, lawmakers in support of A.B. 1709 have already admitted that this bill is likely to follow the same path as other recent "child safety" laws that were struck down or blocked in court for First Amendment and privacy reasons. With A.B. 1709, taxpayers are being asked to hand over a blank check for millions in legal fees to defend a law that is unconstitutional on its face.

Californians: Act Now to Kill This Bill

A.B. 1709 is not an inevitability, as some supporters want you to believe. But we need to act now to support our youth and their right to participate in online public life.

Your representatives could vote on A.B. 1709 as soon as next week. If you’re a Californian, email your legislators now and tell them to vote NO on AB 1709.

Take action

Tell Your Representative to OPPOSE A.B. 1709

EFF calls on the Kuwaiti government to immediately release journalist Ahmed Shihab-Eldin. An award-winning journalist and television host who worked for Al Jazeera for many years, Shihab-Eldin—a dual American-Kuwaiti citizen—was arrested in Kuwait on March 3 while visiting family. The Committee to Protect Journalists (CPJ) reported yesterday that it is believed he has been charged with spreading false information, harming national security, and misusing his mobile phone.

According to the Guardian, Shihab-Eldin published footage of a U.S. Air Force F-15 E Strike Eagle crash, and posted to his Substack about the incident, noting that video circulating online showed local residents assisting the crash survivors. 

Kuwait is one of several countries that has recently cracked down on reporting amidst the ongoing war. Kuwait’s Ministry of Interior posted on X on March 3—the same day Shihab-Eldin was arrested—warning people in the country “not to photograph or publish any clips or information related to missiles or relevant locations.” Earlier this month, the UN Office of the High Commissioner for Human Rights (OHCHR) highlighted a new decree in Kuwait banning the circulation of reports that seek to “undermine the prestige of the military” or erode public trust in it. 

As reported by local media, the decree states that “those who intentionally publish statements or news or circulate false reports and rumors about military authorities resulting in weakening the trust in them and their morale, in addition to undermining their prestige, are punishable by three to 10 years in jail and a fine between KD 5,000 and 10,000.” The decree also imposes a penalty ranging from seven years to life imprisonment for “authorized people who cause financial loss or damage to the military authorities while carrying out a transaction, operation, project or case or obtaining any profit from such deals.”

In contrast to neighboring Gulf states, Kuwait has historically allowed the press to operate with relative freedom, and even introduced a law in 2020 protecting the right to access information. In practice, however, the government exercises considerable control over the media. Furthermore, there are several laws, including cybercrime legislation introduced in 2016, that restrict freedom of expression.

EFF is deeply concerned that Ahmed has not been seen nor heard from in nearly six weeks. We call on the government of Kuwait to immediately release Ahmed Shihab-Eldin.