Dating app fraud in 2026: what we're seeing across Hinge, Tinder, and Bumble
Every month we run thousands of profile scans through TruthHound. The patterns aren't even — some platforms are visibly tightening their safety nets, others are losing ground. Here is what the 2026 data actually shows, written without sales-pitch hedging.
The headline: scams are migrating, not declining
Total volume of romance-fraud reports in the UK and US continued to climb through 2025, and the first quarter of 2026 is on track to exceed last year. The shape of the threat has changed though. The classic "Nigerian prince" email is essentially gone. The standard catfish — stolen photos, fake LinkedIn — is still common but easier to catch. What's grown sharply is the polished, multi-week pig butchering operation that ends in a fake crypto platform.
Hinge: the surprise loser
Hinge had the cleanest reputation among the major dating apps for most of 2024. Through 2025, the picture changed. Our data shows a roughly 40% increase in flagged profiles on Hinge year-over-year, more than any other major platform. Why: pig butchering operations have learned that Hinge's depth-oriented design is better for their playbook than Tinder's. A polished prompt-based profile, two months of slow rapport, then the investment pivot — Hinge users are conditioned to expect long, thoughtful conversations, which is exactly the medium these scams need.
Tinder: roughly stable, scams getting cruder
Tinder's safety improvements through 2024-2025 made a real dent. Their photo-verification system catches the laziest catfish before the first message. What's left on Tinder skews toward the less sophisticated end: WhatsApp pivots within three messages, low-effort profiles, and faster cycles. If you treat the WhatsApp pivot as a hard red flag and walk away, you avoid most of what's currently running on Tinder.
Bumble: the quiet shift to "Bizz"
Bumble's main app didn't show a big change, but Bumble Bizz did. Networking-as-cover scams — where the pitch starts as professional mentorship and escalates to "I'll show you a private investment opportunity" — are now the dominant fraud type on that surface. The "women message first" structure pushed scammers toward a different vector, not away from the platform.
What's new: the AI-photo crossover
Through 2024, AI-generated profile photos were rare on real dating apps because the platforms' verification systems caught them. Through 2025, three things changed: diffusion models stopped producing the obvious tells, scam operations started running photos through compression-and-rephotograph pipelines to defeat detection, and platform safety teams shifted resources to other threats. Our scan data shows AI-generated faces in roughly 8% of flagged profiles in Q1 2026, up from under 2% the year before.
What worked
Apple's iMessage Contact Key Verification, rolled out broadly in 2025, made the "wrong number" iMessage approach significantly harder. WhatsApp's tighter restrictions on bulk messaging from new accounts has slowed (but not stopped) the standard pivot playbook. Hinge's video-prompt feature, where users can post a short verified clip, has reduced catfishing in the subset of profiles that use it — but adoption is still under 20%.
What didn't work
Mandatory ID verification on dating apps is, in 2026 as in 2024, mostly theatre. Scammers buy KYC packages on the open market for under $50. The bottleneck is not "do you have a passport photo to upload" — it's "do you have organised crime budget", and the people running these scams do.
Voluntary photo verification (the blue tick) helps some, but its absence isn't proof of fraud — many real users skip it. Conversely, its presence isn't proof of authenticity — verified accounts are sometimes hacked or sold.
What we expect for the rest of 2026
Three things, in order of confidence:
Voice-cloned scam calls will become routine. The "your daughter has lost her phone" voice note is still mostly a 2025 phenomenon in our data, but it's growing fast. By Q4, expect it to be a baseline threat across WhatsApp and SMS.
Live-video deepfakes will appear in significant numbers. Real-time face-swapping technology hit the consumer threshold in late 2025. We're already seeing the first cases. Within a year, "I did a video call with them" will no longer be the gold-standard proof of authenticity it currently is.
Platforms will not solve this on their own. They cannot. The economics of romance fraud — high payout, low marginal cost, global jurisdictional gaps — are stronger than any single platform's safety budget. The defensive layer that matters is the user's, supplemented by external tools.
What this means for users
Treat dating apps as a starting point, not an endpoint. The verification work — checking the photos, checking the footprint, doing the live video, listening for the early money signal — happens off-platform. That work is what TruthHound is built to make faster and clearer.
We will not solve dating-app fraud in 2026. Nobody will. But the gap between "I had a feeling" and "I have an evidence pack" is the gap that matters, and that gap is closing.