The Law That Built the Internet Turns 30 — and Everyone Wants to Break It Differently

NewsOnScale Staff

October 13, 2026

Thirty years ago, Congress inserted 26 words into a telecommunications bill that would quietly become the load-bearing wall of the modern internet. Section 230 of the Communications Decency Act declared that online platforms could not be treated as the publisher or speaker of content their users posted. It was a practical legal fix for a chaotic new medium. It became, depending on who you ask, either the greatest gift to free expression in the digital age or the legal mechanism that allowed surveillance capitalism, algorithmic radicalization, and AI-generated disinformation to scale without consequence.

The Senate Commerce Committee's hearing marking the law's anniversary did not resolve that dispute. It did, however, clarify how many different directions the pressure is now coming from — and why the next version of this fight will be harder to ignore than the last.

## A Coalition of Critics With Incompatible Goals

The political coalition pushing to reform or repeal Section 230 has always been ideologically incoherent, and the anniversary hearing made that tension visible again. Conservative legislators argue the law is being exploited by large platforms to suppress right-leaning speech without accountability. Progressive critics contend the same immunity shield allows those platforms to profit from harassment, health misinformation, and content that harms children. Both sides want change. Neither agrees on what change would look like in practice.

What's new in this cycle is the growing seriousness of the design-defect argument — the claim that a platform's liability exposure shouldn't hinge on who created a piece of content, but on how the platform was engineered to surface it. The Massachusetts Supreme Judicial Court recently declined to extend 230 immunity to claims targeting social media design choices rather than specific posts. That's a meaningful crack in the wall. Courts finding that algorithmic amplification is a design choice, not neutral hosting, could reshape liability doctrine without Congress doing anything at all.

## AI Changes the Math Entirely

The more consequential long-term pressure on Section 230 may not come from either political flank but from the architecture of AI itself. The law's original logic assumed a clear distinction: platforms were conduits, users were speakers. AI-generated content collapses that distinction. When a large language model produces text in response to a user prompt and a platform delivers that output, who is the speaker? The user who typed the query? The company that trained the model? The platform that integrated it?

Legal scholars and advocacy organizations are increasingly arguing that the passive-host framework simply does not map onto systems that synthesize, generate, and personalize content at scale. This is not a fringe position. It is showing up in legislative proposals, in academic literature, and now in Senate testimony.

## What Platform Suppression Actually Looks Like at Scale

For readers following the platform suppression beat specifically, the 230 anniversary hearing matters for a reason that often gets lost in the abstract legal debate: immunity from liability and power over speech are not the same thing, but they interact constantly. The current legal framework gives platforms enormous discretion to moderate — or not moderate — with limited external accountability in either direction. A platform can suppress a political account, amplify a conspiracy theory, or quietly throttle a category of content, and 230 has historically provided cover for all of it.

Reforming that framework without making it worse requires a level of policy precision that Congress has not yet demonstrated it possesses. Stripping immunity broadly risks chilling legitimate moderation and flooding courts with frivolous claims. Leaving the law unchanged in the face of AI-generated content at scale creates a different set of harms that the original authors of the statute did not envision and could not have.

## The Window Is Narrowing

Thirty years is a long time for any law governing technology to remain structurally intact. The evidence from this hearing cycle suggests that the next major disruption — a high-profile AI harm case, a court ruling on algorithmic design liability, a foreign-influence operation tied to a domestic AI tool — could shift the political calculus quickly. When that happens, the shape of whatever replaces 230 will depend almost entirely on which coalition got there first with a workable framework.

That work is not done. And the clock is running.