Two senators have set a deadline of September 1 for TikTok to clarify its safety evaluations.

Two senators have set a deadline of September 1 for TikTok to clarify its safety evaluations.

      Two US senators have reached out to TikTok concerning an experiment that disabled a safety feature for approximately 15 million American users. Their correspondence spans four pages and includes 13 inquiries. They have set a deadline of September 1 for the company to respond, labeling the test's decision as "depraved."

      Republican Marsha Blackburn from Tennessee and Democrat Richard Blumenthal from Connecticut sent the letter on Wednesday, as reported by Bloomberg. It is directed to TikTok's CEO Shou Chew and Adam Presser, who oversees the company's US division. Both senators are co-authors of the Kids Online Safety Act, indicating that this is a request from those crafting the legislation intended to regulate the company.

      The senators' inquiries are precise. They are demanding the names of all employees who were informed about the experiment, an explanation for including minors in the test, and the timeline of when executives became aware. Additionally, they are requesting a "complete, unredacted version" of the internal document that outlines the experiment.

      One inquiry goes beyond this particular safety test, asking for a comprehensive list of all algorithmic experiments conducted by TikTok in the United States that involved "withholding, disabling, delaying, or reducing a safety feature," along with the number of users affected, including how many were minors.

      The letter begins by referencing "disgusting new reports that TikTok knowingly withheld a critical safety measure for millions of American users," including children. The purported objective of the experiment was to assess whether protecting users would negatively impact profits. The findings "raise serious questions about TikTok’s repeated assurances to Congress, parents, and the American public" regarding prioritizing the welfare of young people over profit, the senators noted.

      The background document

      The experiment came to light through a confidential TikTok document from 2023, which was disclosed during litigation against major social media firms and was sealed by a court. Bloomberg Businessweek published details on August 4. The document indicated that TikTok intentionally kept a safety measure from 15 million US users.

      According to the document, TikTok disabled an algorithmic safeguard for 10% of its US user base, utilizing them as a control group. This safeguard was meant to disrupt echo chambers of harmful content. At that time, 10% represented roughly 15 million individuals, including Chase Nasca, a 16-year-old from Bayport, New York, who was randomly selected on January 25, 2022, and tragically died by suicide within a month.

      The document clearly states the reason why his account differed: “TikTok’s filter bubble prevention strategies did not take effect on this user by design,” highlighting that the decision was based on a "delicate balance" between safety and measuring the impact on daily active users (DAU) and core metrics.

      TikTok's response

      TikTok did not provide a comment regarding the letter but did issue a statement for the Businessweek investigation. A company spokesperson stated that TikTok is “deeply committed to the safety and well-being of users,” particularly teenagers, adding, “Our hearts break for any family that experienced a tragic loss.”

      The spokesperson also mentioned the company's ongoing financial investments in Trust and Safety, emphasizing detection systems and enforcement teams tasked with removing content that violates its community guidelines.

      The senators contend that the timing of the experiment exacerbates the situation, noting that Congress has been expressing concerns since October 2021 about algorithms pushing young users towards harmful content. Since the experiment was initiated after that date, TikTok was therefore aware of the impact of its recommendation systems on children, they argued.

      The relevant legislation

      Blackburn and Blumenthal are advocating for the expedited passage of their bill, the Kids Online Safety Act, which would mandate that platforms prioritize child safety over profits. The bill was initially stalled in the House in 2024 but has since been reintroduced. On August 5, the Senate Commerce Committee voted to advance this revived version.

      The Senate draft retains a duty-of-care clause requiring companies to ensure reasonable protection against mental health harms to minors, such as anxiety, depression, and compulsive usage. Blackburn believes this provision would discourage future experiments, as safety would need to be integrated from the beginning.

      Conversely, the House has opted for a different approach, which is at the heart of the ongoing dispute. Its package addressing online safety for children passed in June and also encompasses AI chatbots and video games, eliminating the duty-of-care provision due to First Amendment concerns. This issue is currently being debated in the courts, where a judge in California recently ruled on First Amendment matters related to addictive feeds.

      Florida Republican Gus Bilirakis, who led the House bill, has had a separate history concerning this matter. During a March 2023 hearing, shortly after the confidential document was drafted, he questioned Chew about Chase Nasca and asserted that his technology was leading to fatalities. Chew responded by stating that the company takes such issues seriously and offers mental health

Other articles

Android 17 prefers to slow down a single problematic app instead of affecting the performance of your entire phone. Android 17 prefers to slow down a single problematic app instead of affecting the performance of your entire phone. Android 17 implements tougher memory limits for individual apps, which can either slow down or shut down apps that are using excessive memory. This provides Google with a fresh method to safeguard overall phone performance before a single memory-intensive app leads to broader issues. Cinemas, educational institutions, and ICE are all prohibiting Meta smart glasses. Cinemas, educational institutions, and ICE are all prohibiting Meta smart glasses. Cinemas, US educational institutions, and ICE are all imposing limitations on Meta smart glasses, each for reasons that are unrelated. The regulation in cinemas goes back to 2014. OpenAI aims to track AI misuse without requiring customers to submit their data. OpenAI aims to track AI misuse without requiring customers to submit their data. OpenAI's Private Safety Processing aims to identify abuse across various interactions while maintaining Zero Data Retention, directly contrasting with Anthropic's 30-day data retention policy. Octopus Charge consolidates 210,000 US charging stations into a single app, excluding Tesla, Ionna, and Electrify America. Octopus Charge consolidates 210,000 US charging stations into a single app, excluding Tesla, Ionna, and Electrify America. Octopus Energy has introduced Octopus Charge in North America, featuring 210,000 chargers. In Europe, the service boasts 1.5 million chargers, which is attributed to regulatory requirements. Cinemas, schools, and ICE are all prohibiting the use of Meta smart glasses. Cinemas, schools, and ICE are all prohibiting the use of Meta smart glasses. Cinemas, schools in the US, and ICE are all placing limitations on Meta smart glasses, each for reasons that are unrelated. The rule concerning cinemas has been in place since 2014. Milan's Domyn secures over $1 billion in what is Europe's largest funding round for AI, primarily through debt financing. Milan's Domyn secures over $1 billion in what is Europe's largest funding round for AI, primarily through debt financing. Domyn, located in Milan, has secured approximately $1.1 billion, with about 90% of that being debt, to enhance the Colosseum supercomputer it developed in collaboration with Nvidia in southern Italy.

Two senators have set a deadline of September 1 for TikTok to clarify its safety evaluations.

Two senators set a deadline of 1 September for TikTok to respond to 13 inquiries regarding the safety experiment conducted on 15 million users in the US.