TikTok's safety experiment: senators demand answers
Two US senators have turned one of the year's darkest platform-safety stories into a regulatory deadline: TikTok has until September 1 to explain an experiment that removed a mental-health safeguard for millions of users — including a 16-year-old who later died by suicide.
Senators Marsha Blackburn (R-TN) and Richard Blumenthal (D-CT) have given TikTok a September 1 deadline to answer 13 questions about an experiment that disabled an algorithmic safeguard for 10% of US users — roughly 15 million people — calling the decision "depraved" in a letter to CEO Shou Chew and US business chief Adam Presser. Bloomberg first reported that TikTok intentionally switched off the safeguard, which was designed to prevent users from being overwhelmed by harmful content, to test whether the protection made the app less engaging. One of the users swept into the control group was 16-year-old Chase Nasca, whose feed served him thousands of videos about sadness, suicide, and loneliness up until his death by suicide.
The senators' questions go beyond this single case: they want the names of every employee who knew about the experiment, an explanation of why minors were allowed into the control group, and a full list of every algorithmic experiment in the US in which TikTok disabled or delayed a safety feature. That last demand matters most. It converts an individual tragedy into a request for a complete accounting of when the company has treated safety guardrails as test variables rather than protections — the pattern that makes the difference between tuning a recommender and experimenting on the people inside it.
The uncomfortable reality is that this is exactly how engagement-optimized platforms are built: A/B tests run against retention and watch-time, with safety as the thing most easily traded away. Fifteen million people did not consent to being a control group for "does harm keep them scrolling," and the fact that a child was among them is what turns this from a design critique into a governance story. TikTok has not yet responded publicly. The September 1 reply will show whether the company can produce a full inventory of its safety-feature experiments — and it sets the stage for the FTC and Congress to decide whether algorithm testing this invasive needs a consent regime of its own.
What to watch: TikTok's answer to the letter, and whether the disclosure request forces other platforms to admit how often safety features get A/B-tested.
If a platform can run experiments on how much harmful content users can tolerate, what does informed consent even mean? Tell us in the comments.
