Senate Report Says Tinder Went Years Without Updating Most of Its Scam Detection Models

Most of Tinder's machine learning models for catching scammers and other bad actors had not been updated in years as of 2025, according to a report released October 13 by Sen. Maggie Hassan (D-N.H.), ranking member of the Joint Economic Committee. The report, titled "Outmatched: Tinder, Scammers, and Match Group's Efforts to Catch Up," closes a yearlong investigation that Hassan opened with Sen. Marsha Blackburn (R-Tenn.) in September 2025. Match Group provided more than 670 pages of internal documents and briefed committee staff.

The report's central finding is that scammers' reach on Tinder grew while detection stood still. Globally, the number of profile views received by the median scammer rose from fewer than 70 in May 2024 to more than 500 by December 2025, an increase of more than 700%. Tinder's own internal documents, cited in the report, said scam and other abuse accounts "usually have a higher exposure / impression rate than regular accounts." In the U.S., about 1.5 million spam, scam and other inauthentic accounts on Tinder and Hinge matched with at least one user in the first nine months of 2025, and nearly 500,000 of them exchanged contact information.

The report does not settle why scammers got more exposure. Match attributed it to scammer behavior, such as swiping right on nearly every profile, and said it does not optimize for impressions. The report notes that Match did not identify any changes to Tinder's recommender to limit scammers' reach before they were caught. Match told staff its models are retrained every year or every couple of years. Experts the committee consulted said scam detection needs retraining at least monthly.

The report also credits Match with changes. Tinder's Face Check, which requires new users to complete a video selfie, went nationwide in the U.S. in early December 2025. Match told the committee that interactions with bad actors on Tinder, a figure that combines scams and spam, fell by more than 70% within a month and stayed down for at least seven months. Match also deployed a new scam detection model in the first half of 2026 and added 169 automated enforcement rules and improved 108 others in that period. On Hinge, Face Check reportedly cut interactions with bad actors by 20% to 30% in key markets.

"Match Group made progress in combating scammers, but it shouldn't have taken so long," Hassan said. Yoel Roth, Match Group's senior vice president of trust and safety, told NBC News the report "mischaracterizes both the historical scale of our investment and the work already completed." The report makes no formal recommendations. It lands weeks after Congress sent the Romance Scam Prevention Act to the President. That bill will require dating apps to warn users who messaged an account later banned for fraud.

Why it matters: This is the first time a congressional report has put a dating company's internal fraud metrics on the record, and the measure it used was model freshness, not policy. Regulators now have a benchmark: how often you retrain, how far scammers reach before you catch them, and whether verification is mandatory. Every operator, not just Match, should expect to be asked those questions, and should know its own answers before the FTC starts enforcing the scam notification law.

Sources: "Outmatched" report (JEC Minority, PDF) · Sen. Hassan press release · NBC News · Boston Globe