US senators demand TikTok explain algorithmic safeguard trial
Lawmakers have given the social media giant until 1 September to answer 13 questions regarding an experiment that withheld a safety feature from 15 million users.
US Senators Marsha Blackburn and Richard Blumenthal have formally requested clarification from TikTok regarding a recent algorithmic trial that affected a significant portion of its American user base. The bipartisan pair sent a letter to TikTok CEO Shou Chew and Adam Presser, the CEO of the company’s US business, demanding answers to 13 specific questions about the experiment.
The senators described the company’s decision to run the trial as “depraved.” According to reporting by Bloomberg, TikTok intentionally disabled an algorithmic safeguard for 10 per cent of US users, creating a control group of approximately 15 million people. The safeguard was originally designed to prevent users from being overwhelmed by harmful content, but the company wished to determine whether the feature made the app less engaging.
The inquiry follows the death of 16-year-old Chase Nasca, who was part of the control group. Nasca died by suicide after his algorithm reportedly fed him thousands of videos about sadness, suicide, and loneliness. The senators are seeking to establish the company’s liability through their questions, specifically asking why TikTok allowed minors to be included in the control group.
The letter also requests the names of every employee who knew about the experiment. Additionally, the senators are seeking a comprehensive list of every algorithmic experiment in the US where TikTok disabled or delayed a safety feature. This broad inquiry aims to uncover the extent to which the company has prioritised engagement metrics over user safety in its algorithmic testing.
TikTok has been given until 1 September to respond to the inquiries. As of the time of reporting, the company had not immediately responded to requests for comment. The outcome of this correspondence may influence broader regulatory discussions regarding how social media platforms manage algorithmic risks for their users.
