AI fraud tools draw warnings from researchers and firms (opens in new tab)

Researchers and security firms described new evidence that artificial intelligence is being used in scams, impersonation and financial promotion, as researchers develop ways to measure and detect scam progression. An arXiv paper introduced PreScam, a benchmark built from 177,989 user-submitted scam reports and organized into 11,573 conversational scam instances across 20 scam categories to test whether models can predict scam progression from early exchanges. The reports cover consumer warnin...