The House of Representatives passed the KIDS Act (H.R. 7757) on June 29, 2026 by a vote of 267-117, bundling fourteen bills — a revised Kids Online Safety Act (KOSA), an expansion of COPPA to teenagers, mandatory age verification for pornographic platforms, AI-chatbot disclosure requirements, and new limits on minors' access to disappearing messages — into a single package headed for the Senate. The bipartisan margin is real, but it papers over the fact that the House only got there by cutting the single most contested provision in online child-safety law: KOSA's "duty of care."
What Actually Changed
The original Senate KOSA (S. 1748) would have imposed a duty of care requiring platforms to design products that prevent and mitigate harms like anxiety, depression, and addictive engagement — a standard critics warned was vague enough to make platforms over-remove any content a regulator might later call harmful. The House-passed version explicitly disclaims that standard: the bill states the relevant provision "should not be construed to impose a duty of care on covered platforms," according to a client alert from Pillsbury Winthrop. That single edit is why the House could assemble 267 votes — and why Senate KOSA sponsors have reportedly called the House bill dead on arrival in the upper chamber.
What's left is more concrete and more defensible. The bill expands the Children's Online Privacy Protection Act beyond its current under-13 scope: it redefines "child" as under 14 and creates a new "teen" category for ages 14-17, extending parental-consent-style obligations and giving teens rights to access and delete their own data. It requires AI chatbot operators to clearly disclose that a system is artificial, bars chatbots from falsely claiming professional licensure, and requires them to surface crisis resources when a minor raises self-harm. It restricts minors' use of disappearing-message features. And it mandates age verification for platforms hosting content obscene to minors.
The Case For It, Stated Fairly
The strongest argument for this package isn't sentimental — it's evidentiary. COPPA has covered under-13s since 1998 while an entire cohort of 13-to-17-year-olds using the same addictive design patterns, the same data-harvesting ad models, and now the same AI companion apps has had no equivalent floor. Requiring an AI chatbot to disclose it isn't human, or to hand a suicidal teenager a crisis line instead of role-playing through the conversation, is a narrow, disclosure-based rule that doesn't ask a platform to predict which content causes "harm" — it asks for a label and a safety valve. That's a meaningfully different, more tailorable ask than the duty-of-care standard the House dropped, and it's why even bill skeptics distinguish the AI-disclosure and COPPA-teen provisions from the age-verification mandate.
Where It Breaks Down
Age verification is where the bill's proportionality argument collapses. The Electronic Frontier Foundation puts the core problem plainly: "there is no way to determine a user's age online that is both privacy protective and accurate." Every viable method — government ID upload, biometric face-scanning, or behavioral inference — requires linking an adult's offline identity to their online reading habits, then storing that link somewhere a breach can expose it. EFF also flags that KOSA's retained content-policy language, covering gambling, alcohol, and cannabis-adjacent speech, pressures platforms to over-filter topics teenagers legitimately search for help on, like a parent's gambling problem or a friend's addiction.
The legal backdrop makes age verification harder to challenge than it would have been two years ago. The Supreme Court's June 27, 2025 ruling in Free Speech Coalition v. Paxton upheld Texas's age-verification law for pornographic websites 6-3, with Justice Thomas writing that the mandate "only incidentally burdens the protected speech of adults" and applying intermediate rather than strict scrutiny — a marked departure from the strict-scrutiny approach the Court took in 2004's Ashcroft v. ACLU. That precedent gives Congress a green light it didn't have before to write federal age-gating into law, but a green light on constitutionality isn't the same as evidence the mandate will work as intended or that the data it generates will stay secure.
The Senate's Real Choice
The Senate now faces a genuine trade, not a rubber-stamp vote. It can restore KOSA's duty of care and lose the bipartisan coalition that got the House to 267 votes, hold the House's narrower disclosure-and-verification package as-is, or do what EFF is actually asking for — drop age verification, keep the disclosure and COPPA-teen provisions, and pair them with a comprehensive federal privacy law that limits behavioral tracking of everyone, not just minors. The last option is the one that would most directly address the underlying harm without building a national identity-verification infrastructure for the open internet. Congress has now had every element of this debate — a First Amendment ruling on point, a House-passed text, and organized opposition — on the table simultaneously for the first time. There's no more research to wait for; the Senate's version will be a statement of what it actually values.