Anthropic’s September 10, 2026 report dives into widespread AI abuse. One standout case: a network of over 20 Chinese dating apps where for every real profile, there were about three AI personas. These digital chat partners kept up conversations, built fake 'relationships,' and squeezed money out of users.

The Vibe-Hacking Economy

Vibe-hacking is all about cashing in on attention and trust by faking intimacy. Users think they’re paying for real human connection and emotion, but more often than not, they’re just chatting with a synthetic persona. Spinning up tons of these fake accounts creates a steady stream of conversations, way easier to scale than hiring real people, and turns users’ time and trust into revenue.

What Else the Report Uncovered

The dating app networks are just one of dozens of cases Anthropic breaks down. The report details how AI is being used for cyberattacks, data theft, scams, and even political manipulation. The big trend: scammers are shifting from one-off money grabs to systematically hijacking attention and access. Getting the 'keys' to data and accounts lets them quietly monetize victims over the long haul.

Why This Matters

The market’s filling up fast with convincing AI personas, and the old trust signals don’t cut it anymore. For platforms, it’s a challenge of verifying users and spotting these synthetic networks. For users, it’s about staying sharp—because emotional engagement could just be a setup to drain your wallet.