Credit: Jernej Furman from Slovenia
Two US senators have written to TikTok about an experiment that switched off a safety feature for roughly 15 million American users. Their letter runs to four pages and sets out 13 questions. It gives the company until 1 September to reply. The senators call the decision to run the test “depraved”.
Republican Marsha Blackburn of Tennessee and Connecticut Democrat Richard Blumenthal sent it on Wednesday, Bloomberg reported. It is addressed to TikTok chief executive Shou Chew and to Adam Presser, who runs the company’s US spinoff. The pair co-authored the Kids Online Safety Act. That makes this a demand from the people drafting the law meant to govern the company.
What the senators are asking for
The questions are specific. The senators want the names of every employee who was told about the experiment. They want an explanation of why the company allowed minors into the test. They want to know when executives learned of it. They also want a “complete, unredacted version” of the internal document describing it.
One request reaches far beyond this one safety experiment. The pair asked for a list of every algorithmic experiment TikTok has run in the United States where it “withheld, disabled, delayed, or reduced a safety feature”. They want the number of users caught in each one. They also want to know how many of those were minors.
The letter opens by citing “disgusting new reports that TikTok knowingly withheld a critical safety measure for millions of American users”, children included. The stated purpose of the test was to find out whether protecting people would dent the bottom line. The revelations “raise serious questions about TikTok’s repeated assurances to Congress, parents and the American public” about putting young people ahead of profit, the senators wrote.
The document behind the letter
The experiment surfaced in a confidential TikTok document from 2023. Lawyers handed it over in litigation against the largest social media companies, and a court placed it under seal. Bloomberg Businessweek reported its contents on 4 August. TikTok had kept a safety fix from 15 million US users by design, the document showed.
TikTok turned off an algorithmic safeguard for 10% of its US users, the document says, making them a control group. The safeguard existed to break up echo chambers of harmful content. At the time, 10% came to about 15 million people. One was Chase Nasca, a 16-year-old from Bayport, New York. He was selected at random on 25 January 2022 and died by suicide within a month.
The document is blunt about why his account differed. “TikTok’s filter bubble prevention strategies did not take effect on this user by design,” it states. It describes the choice as “a delicate balance across safety and the ability to measure impact on DAU (daily active users) and core metrics”.
What TikTok has said
TikTok did not respond to a request for comment on the letter. It gave a statement for the Businessweek investigation. The company is “deeply committed to the safety and well-being of users”, especially teenagers, a spokesperson said then. “Our hearts break for any family that experienced a tragic loss,” the spokesperson added.
The company also pointed to its spending. It keeps investing in Trust and Safety, the spokesperson said, including detection systems and enforcement teams that remove content breaching its community guidelines.
The senators argue the timing makes it worse. Congress had been raising concerns about the algorithm pushing young users towards harmful content since October 2021. The experiment began after that. TikTok was therefore on notice about how its recommendation systems affected children, they wrote.
The bill this feeds into
Blackburn and Blumenthal want their bill fast-tracked. The Kids Online Safety Act would force platforms to put child safety above profit. It died in the House in 2024, then returned this session. On 5 August, the day after the Businessweek story ran, the Senate Commerce Committee voted to advance the revived version.
The Senate text keeps a duty-of-care provision. That clause would require companies to take reasonable care to prevent mental health harms to minors, including anxiety, depression and compulsive use. Blackburn argues it would deter future experiments, because safety would have to be designed in from the start.
The House took a different route, and that difference is the whole fight. Its kids online safety package passed in June and covers AI chatbots and video games too. It strips the duty-of-care provision out over First Amendment concerns. The same question is live in the courts, where a California judge recently issued a First Amendment ruling on addictive feeds.
Florida Republican Gus Bilirakis, who championed the House bill, has his own history here. At a March 2023 hearing, weeks after the confidential document was written, he asked Chew about Chase Nasca and told him his technology was leading to death. Chew replied that the company takes these issues very seriously. It offers mental health resources to users who search for suicide content, he said.
The pressure elsewhere
The letter lands on a company already fighting on several fronts. Roughly 2,400 addiction lawsuits are proceeding after the Ninth Circuit declined to shut them down. TikTok has settled repeatedly rather than let a teen case reach a jury.
European regulators are running parallel processes, and they have their own document powers. Brussels has charged TikTok over children’s privacy under the Digital Services Act. Ofcom opened a child safety investigation in July. A congressional letter carries no subpoena power, so the company can decline it.
What TikTok answers by 1 September, and what it leaves out, will be read closely on both sides of the Atlantic.
Get the TNW newsletter
Get the most important tech news in your inbox each week.
