Earlier in October, U.S. lawmakers revealed how users of dating apps were being exposed to serious financial and physical risks.
asserted
users → reveal → risks
They pinned this to the outdated AI stack in these applications.
asserted
They → pin → applications
In an investigation centred around the Match Group, the lawmakers found that the algorithm matched U.S. users of dating apps like Tinder and Hinge with scammers.
asserted
algorithm → centre → scammers
The report, titled ‘Outmatched: Tinder, Scammers, and Match Group’s Efforts to Catch Up’, highlighted that users connected with nearly 1.5 million scammers on these two popular dating applications in the first nine months of 2025.
asserted
users → title → 2025
At the core of these disturbing findings were outdated machine learning models installed to detect bad actors.
asserted
models → instal → actors
These hadn’t been updated in years, per the report.
asserted
These → update → report
This information came from internal documents that the Match Group shared with the committee, led by senator Maggie Hassan.
asserted
Group → come → Hassan
As the accuracy of scam detection dropped, Tinder began depending on other measures that it knew were often ineffective.
asserted
it → drop → measures
“In fact, Match Group knew that Tinder was more likely to show a profile if it belonged to a scammer or other bad actor rather than a regular user,” stated an official press release about the report.
asserted
release → know → report
However, the investigation acknowledged that the Match Group later put in place new safety measures, such as an AI model in Tinder this year that was designed to detect scammers and bad actors.
asserted
that → acknowledge → scammers
The report included statements from victims who described how romance scammers cheated them out of tens of thousands of dollars after building a relationship with them on the dating apps.
asserted
scammers → include → apps
A common trend was to cite a major emergency in order to pressure the victim into sending over money.
asserted
trend → cite → money
Using the victims’ phone number and/or email ID, the scammers would proceed to harass them online until the victim sometimes paid them in the hopes of de-escalating their aggression.
asserted
victim → use → aggression
Another type of scam involved the same malicious user operating as different people on the same dating app, to ensnare more victims.
asserted
user → involve → victims
The report pointed to an inconsistent dating app AI strategy, where patchy machine learning systems led to heavy caseloads and “overwhelmed” human moderators who were unable to focus on visibly risky accounts.
asserted
who → point → accounts
Per a document from the Match Group itself, Tinder could have better tackled dangerous accounts by building better AI signals or reducing its reliance on manual rules.
uncertain
Tinder → tackle → rules
“In practice, these moderation lapses meant that bad actors that reached moderation were not immediately banned and instead may have continued to interact with and harm legitimate users until the platform gathered additional evidence,” explained the report.
uncertain
report → mean → evidence
A bouquet of dating apps come under the Match Group, including Tinder, Hinge, Match (formerly Match.com), Plenty of Fish, and more such offerings catering to users in diverse countries and cultural communities.
asserted
bouquet → come → countries
In other words, a security lapse in one app comes with serious implications for the users of other Match Group-owned apps.
asserted
lapse → come → apps
Tinder and Hinge are popular in India as well, claiming the top ranks in terms of downloads via app marketplaces.
asserted
Tinder → claim → marketplaces
According to the Match Group, Hinge is expected to reach $1 billion in revenue in 2027, while Tinder is expanding its in-person events that are already live in the U.S. and Europe.
uncertain
that → accord → U.S.
Indian users often register with apps that suggest profiles and tout AI-powered matchmaking as a selling point.
asserted
that → register → point
However, as the U.S. committee report noted, cutting-edge AI tools require constant upkeep, testing, and upgrades in order to ensure customers remain shielded from emerging threats.
asserted
customers → note → threats
Regulators must also play a larger role in monitoring dating app services.
asserted
Regulators → play → services
How can users avoid scammers on dating apps?
asserted
users → avoid → apps
Dating app users should first vet the platforms they register with, making sure they agree with how data is collected and processed.
asserted
data → vet → platforms
Users should also be aware of the reporting options they have in case an online or offline interaction turns unsafe.
asserted
interaction → have → options
Customers can research the apps’ founders, investors, and brand ambassadors, to make sure there is an alignment of values.
asserted
Customers → research → values
Dating app users should promptly report fraudulent user profiles and cease interacting with them.
asserted
users → report → them
AI-generated user photos and videos, even when posted as a joke, often violate platform policies.
asserted
photos → generate → policies
Another red flag is a dating app user who becomes intimate too quickly or is eager to discuss finances early on.
asserted
who → become → finances
Furthermore, scammers often pressure dating app users to contact them on external sites, so they can view the user’s phone number or harvest other personal details.
asserted
they → pressure → details
Users must withhold personal phone numbers, email IDs, home addresses, private social media channels, and employer information until they meet a match in-person and feel comfortable proceeding.
asserted
they → withhold → person
The Aporia
