TikTok tested a fake safety toggle on thousands of teens and kids
New York’s attorney general alleges TikTok ran experiments that gave teens and children a non-working safety feature, masking what was really happening in feeds and ads.
New York AG Letitia James alleges TikTok ran experiments using a non-functional safety feature called Algo Refresh on thousands of users, including teens and children.
The lawsuit says some users were in a placebo group and believed the tool worked, but their feeds did not change, while others saw a functional version.
The complaint ties these tests to study of time spent in the app and ad sales, and cites concerns from an internal product manager about the placebo undermining algorithm transparency.
TikTok says it regularly tests new features and that the lawsuit presents a false picture of how Algo Refresh works.
Quick read · 2 min
New York is suing TikTok over experiments in which thousands of users, including teens and children, were exposed to a safety feature called Algo Refresh. The complaint alleges a placebo group, where some users believed the tool worked but did not see feed changes, while others did see a functional version. The case centers on concerns about true algorithm transparency and how safety tools are tested and explained to users.
What this means for you: tests like this shape how clearly platforms explain safety tools and what users can expect from them. The outcome could influence future experiments and disclosures. What happens next: the case moves through the court system, with TikTok responding and courts deciding any required changes.
Algo Refresh placebo claims
Algorithm transparency concerns
State-led action against TikTok
New York is suing TikTok over a set of experiments the state says secretly tested a safety tool on teens and children, including a group that received a placebo and believed it was working. The lawsuit, filed by the New York attorney general and unsealed in parts over the past week, centers on a feature called Algo Refresh. The claims paint a picture of testers trying to measure how well a recommendation reset would reduce exposure to risky or unwanted content, while some users didn’t actually see a real change in their feeds.
In plain terms, the state says thousands of users were part of experiments where the tool was supposed to refresh or reset recommendations. Some users were in a placebo group that didn’t actually alter their feed, yet they thought the feature was active. The contrast is that other users did get a fully functioning version of the tool. The gist, according to the complaint, is that a placebo group directly conflicts with the aim of algorithm transparency and control.
The lawsuit connects these tests to broader questions about how the platform studies user experience, including how much time people spend on the app and how much advertising viewers see. A TikTok product manager involved in the Digital Well Being team reportedly raised concerns internally, calling the placebo group a problem for the idea of true algorithm transparency.
TikTok has publicly defended its approach to testing, saying the company often probes new products and features and that the lawsuit misrepresents how Algo Refresh works. The company did not immediately respond to our requests for comment on the latest filing.
01
What Algo Refresh is supposed to do
Algo Refresh is described as a safety feature meant to let users reset their recommendations so the app can show less of certain content. In broader terms, it’s a tool for giving users more control over what the algorithm surfaces. The New York suit, though, argues that what users experienced in tests did not match the stated purpose of real transparency and control.
02
What the lawsuit alleges happened in the tests
The attorney general’s office says the studies involved thousands of users, including minors, and that some participants did not receive an actual safety improvement even though they believed they did. The complaint cites an internal chat in which a product manager reportedly wrote, “This feature is about algorithm transparency and control. A placebo group completely conflicts with that.”
03
Why this matters for everyday users
For parents and frequent TikTok users, this case highlights how experimental features can affect what kids see online and how safe those tools really are. If experiments are not transparent or if placebo groups are used to gauge impact, it becomes harder to trust claims about safety features or content controls. It also raises questions about what data companies collect during testing and how they measure “success” in ways that could influence ads and recommendations.
04
What happens next
The New York case is part of a broader wave of state actions against TikTok over safety, child safety, and the platform’s impact on young users. The attorney general’s office will likely push for findings on whether TikTok misrepresented safety tools and how experiments may have affected minors. For everyday users, the outcome could influence how future features are tested and how clearly platforms disclose experimental changes.
A placebo tool is a feature that looks and acts like a real tool but doesn’t actually change anything. In this case, some users thought Algo Refresh was helping control their feed, but it didn’t alter recommendations.
Will this affect my TikTok right now?
The case is ongoing. It doesn’t change TikTok’s day-to-day features unless a court orders relief or changes to how experiments are conducted.
What should I look for as a parent?
06
What happens next
The legal process will unfold over months as the state builds its case and TikTok responds. We’ll keep you updated on any substantive changes to safety features or how tests are disclosed to users.
New York AG vs TikTok
Algo Refresh safety tool
Algorithm transparency
07
Sources
Official court filings and statements will drive what comes next in this dispute. Keep an eye on updates from the attorney general’s office for the latest.
A lengthy new GTA 6 video circulating online reportedly shows major story spoilers and a nude scene, with the official release still set for November 19.
3 min read
We use cookies to understand how readers use Talk With Tech, so we can make it better. Is that OK? Privacy policy
The Daily Brief
Today's biggest tech stories, in 5 minutes
Every morning, the news that matters from AI, phones, apps and the people shaping tech. Explained in plain English. Free.
One email a day. No spam, unsubscribe anytime. Privacy policy
