Courts Are Cracking Section 230's Armor — and the Platform Economy Is Watching

NewsOnScale Staff

September 15, 2026

For most of the internet's commercial life, Section 230 of the Communications Decency Act functioned less like a law and more like a force field. Platforms were not publishers, courts said, again and again. What users posted was not the platform's problem. The result was an extraordinary accumulation of legal immunity that allowed companies like Meta and Google to scale into global infrastructure with almost no tort exposure.

That arrangement is now under meaningful strain — and not because Congress rewrote anything.

## What Changed, and Why It Matters

The recent verdicts against Meta and Google documented by Tech Policy Press represent something genuinely new in platform litigation: juries and courts holding companies liable not for what appeared on their services, but for the deliberate architectural choices that shaped how users encountered that content. Autoplay. Infinite scroll. Algorithmic amplification calibrated to maximize engagement regardless of psychological cost. These are design decisions, made by engineers and product managers, reviewed by executives, and optimized over years of A/B testing.

The legal theory is precise: Section 230 immunizes platforms as distributors of third-party speech. It does not, plaintiffs argue — and some courts are now agreeing — immunize the engineering decisions that determine whose speech gets amplified, at what hour, to which neurological vulnerabilities, and with what addictive feedback loops attached.

This is not a fringe reading. The Massachusetts Supreme Judicial Court recently joined a growing number of state courts in finding that design-defect claims can survive a 230 motion. The reasoning is consistent: a platform's recommendation engine is a product. Products can be defective. Defective products cause harm. Immunity from defamation claims does not extend to product liability.

## The Accountability Gap This Exposes

What makes this litigation wave significant for anyone tracking platform power is what it reveals about the gap that has existed in accountability infrastructure. Regulatory agencies moved slowly or not at all. Legislative reform stalled in partisan gridlock. Voluntary transparency commitments from platforms produced dashboards that disclosed almost nothing useful about how ranking systems actually operate.

Litigation, for all its inefficiencies, has a discovery process. Plaintiffs' attorneys in these cases have obtained internal research, product roadmaps, and communications that regulators never saw and journalists could not compel. The picture that has emerged — of companies that knew their systems were causing measurable harm to specific user populations and optimized for engagement anyway — is not a picture that platforms chose to share publicly.

## What Platforms Are Saying, and What That Signals

The industry's legal and policy response has been to defend the broad interpretation of 230 as essential not just to their own business models but to the open internet itself. The Cato Institute's position, representative of a libertarian strand of tech policy thought, holds that weakening immunity will chill innovation and burden smaller platforms that lack the legal resources of a Meta or Google.

That argument deserves serious engagement. Poorly drawn design-liability standards could indeed create compliance burdens that entrench incumbents and raise barriers for new entrants. The solution to platform concentration is not a legal environment that only the largest platforms can navigate.

But the argument also functions, in practice, as a reason to never impose accountability for anything. And the internal documents surfaced in litigation suggest that the harm caused by specific design patterns was not an externality platforms failed to foresee — it was a known cost that was accepted because engagement metrics rewarded it.

## The Horizon

The legal landscape will take years to settle. Appeals are pending. Federal courts may draw different lines than state courts. Congress could theoretically act, though its track record on 230 reform is one of loud debate and minimal output.

What has already shifted is the premise. The question is no longer whether platform design choices are outside the scope of legal accountability. Courts have established that they are not. The question now is how precisely that accountability gets defined — and whether the definition will be rigorous enough to reach the specific engineering decisions that caused documented harm, or vague enough to leave the real machinery of algorithmic power untouched.