New York Attorney General Letitia James alleges that TikTok handed thousands of users, teenagers and children among them, a version of its “Algo Refresh” feed-reset tool that did nothing. Users who believed they had switched the feature on saw no change in what the app recommended. The claim comes from an updated complaint that a judge ordered unsealed on Oct. 7, 2026, and TikTok disputes it.
According to the filing, the experiments ran repeatedly in 2023 and were built to measure what a working reset would cost the company in time spent on the app and in advertising revenue.
The Algo Refresh placebo tests described in the unsealed complaint
As Reuters reporting carried by WMBD describes the amended complaint, TikTok picked thousands of users who had opted into safety features such as Algo Refresh and gave some of them a non-functional version. They compared those users against people who received the real tool, tracking time on the platform and ad revenue. The updated complaint, which was filed under seal at the end of August, is the source of the account, and the placebo claims come from that sealed version rather than from the original 2024 pleading, which is public in redacted form.
The filing quotes an internal exchange in which a TikTok digital well-being product manager warned that placebo tests on Algo Refresh were problematic and could mislead users. In a chat message, the employee wrote that “A placebo group completely conflicts with that,” according to the same report. EconoTimes, summarizing the unsealed complaint, adds that the manager’s objection was that a placebo group contradicted the feature’s purpose of giving people more control and transparency over the algorithm.
The “start fresh” promise in the 2024 pleading
The placebo allegation sits on top of an older claim. New York’s original lawsuit, filed on Oct. 8, 2024 in a coalition with 13 other attorneys general, already said TikTok misled the public about the effectiveness of safety tools including “Refresh.” City & State reported at the time that James said the company was “misleading the public about the safety of its platform and harming young people’s mental health.”
The redacted 2024 complaint sets out the mismatch in detail. Paragraph 204 says TikTok billed the feature as “[t]he option to start fresh on TikTok,” and paragraphs 207 and 208 say that would lead consumers to expect a completely new feed. The complaint’s answer is blunt: “But that is not what it does.” It alleges Refresh does not reset personalized ads, was hard to find, and that users “may find that they are quickly back in the same rabbit hole again.”
That complaint does not use the word placebo. The 2024 pleading alleged a tool that worked less thoroughly than advertised, while the unsealed version alleges that for some users it did not work at all.
TikTok’s denial, the statutes and the court record
A TikTok spokesperson told Reuters the company regularly tests features to “validate the experience and understand how they work in the real world,” and that the allegations mischaracterize how Algo Refresh works. The company has broadly denied New York’s claims in court filings and says protecting younger users is a priority. Nothing in the filing has been tested at trial, and every point above is an allegation.
The case, The People of the State of New York, by Letitia James, Attorney General v. TikTok Inc., et al., Index No. 452749/2024, is pending in New York State Supreme Court in New York County. The defendants named in the original complaint are TikTok Inc., TikTok LLC, TikTok U.S. Data Security Inc., TikTok Pte. Ltd., TikTok, Ltd., ByteDance Inc. and ByteDance Ltd. The complaint invokes Executive Law Section 63(12), the state’s anti-fraud statute, along with General Business Law Sections 349 on deceptive practices and 350 on false advertising, and it seeks civil penalties, punitive damages, injunctive relief, restitution and disgorgement. On May 28, 2025, Judge Anar Rathod Patel denied TikTok’s motion to dismiss, which James called a decision she was “pleased with.”
The unsealed filing also repeats a separate allegation first reported by Bloomberg: that TikTok randomly selected users and switched off safety features limiting exposure to harmful content, and that a New York teenage boy included in one such test died by suicide after being shown an “onslaught” of self-harm and suicide material. That claim rests on the same unsealed pleading and carries the same status as the Algo Refresh allegation, which is that a court has yet to rule on it. The suit is among more than two dozen state actions accusing TikTok of designing its platform to addict children and misleading users about safety, and the Reuters report notes that TikTok settled the first of those set for trial, in Alabama, in September.
This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.
More from Morning Overview
- Seven of Earth’s nine planetary boundaries are breached, and all seven are worsening
- The NSA says three phone features should be off whenever you aren’t using them
- Hurricane Hunter radar shows four warning signs that a tropical cyclone is about to strengthen, a University of Miami study found
- Tropical Storm Rachel is dumping up to 12 inches on four Mexican states on its way to major hurricane strength
