A Federal Court Just Blocked Texas's "Harmful Speech" Law

Key Takeaways

• The US 5th Circuit Court of Appeals has blocked a Texas law that would have required websites to filter content deemed "harmful," striking down a state-level mandate for government-directed content moderation.
• The ruling protects platform autonomy and free speech principles against government-compelled filtering — a mechanism that, once in place, is difficult to keep narrowly scoped.
• The definition of "harmful" in laws like this is never as neutral as it sounds: it invariably reflects the political priorities of whoever is doing the defining.
• The decision adds to a growing body of case law pushing back against state governments attempting to conscript private platforms into enforcing their content preferences.

The 5th Circuit Court of Appeals has blocked a Texas law that would have required websites to filter content deemed "harmful," according to reporting by Ars Technica. The ruling strikes down a state-level mandate for government-directed content moderation — and it's the right outcome, even if the law's stated intentions were less sinister than the mechanism it created.

The core problem with laws that compel platforms to filter "harmful" speech isn't that harmful content doesn't exist. It does. The problem is the word "harmful" and who gets to define it. In the context of a state legislature, "harmful" tends to mean "content the current political majority finds objectionable." 

That can mean violent content one day and LGBTQ+ educational material the next. It can mean misinformation about public health or accurate reporting on government conduct. Laws built around vague harm standards don't stay narrow. They expand to cover whatever the government finds inconvenient.

The 5th Circuit's intervention matters because it establishes that states can't simply compel platforms to become instruments of their content preferences. That boundary is important to maintain, particularly as state-level content regulation has accelerated in recent years across both political parties — conservatives pushing to mandate certain kinds of speech, progressives pushing to mandate certain kinds of removal.

“Harmful” Is Always a Political Definition

It's worth sitting with what government-mandated filtering actually requires in practice. A platform told by law to filter "harmful" content needs to know what harmful means. It needs systems capable of applying that definition at scale across millions of pieces of content. 

It needs to err on the side of removal when uncertain — because the legal risk of over-inclusion is lower than the legal risk of leaving something up that a court later decides was harmful. The result is systematic over-moderation of anything that sits near the edge of the definition, which is exactly where the most important speech tends to live: the contested, the uncomfortable, the politically charged.

Young people especially feel the downstream effects of this. The internet is where they organize, find community, access information about their own identities and health, and participate in public life. Platforms compelled to aggressively filter "harmful" content don't just remove extreme material. They remove the gray areas, and the gray areas are often where the conversations that matter most are happening.

The Broader Pattern This Ruling Pushes Back Against

This isn't the first time courts have had to push back against state governments trying to use content mandates to shape the information environment. Texas and Florida have both pursued laws attempting to regulate how platforms moderate content, with courts repeatedly finding that these laws conflict with First Amendment protections for private editorial decisions. 

The Coyote System ruling from the EU Court of Justice last week illustrated that this problem isn't limited to the US: when legal frameworks push platforms to take on liability for content, the predictable result is more removal of lawful speech, not less.

The 5th Circuit's decision doesn't resolve the broader tension between platform power, government authority, and free expression. That tension isn't going away. But it does reinforce something worth holding onto: a government that can compel platforms to filter speech it labels harmful has effectively handed itself a censorship tool. Courts that recognize that (regardless of which party's law is in front of them) are doing important work.

The internet should be a space where kids can be curious, ask questions, find communities, and encounter ideas that challenge them. That doesn't happen when governments get to decide in advance what's too harmful to be seen.