Anthropic Publishes Report on its AI Fraud Detection Efforts
Anthropic has released a new report highlighting how its AI models have been targeted and misused by cybercriminals, as well as the measures it has taken to counter the threats. The Threat Intelligence report outlined multiple cases where its Claude models were implicated in large-scale fraud, extortion, and cybercrime. According to Anthropic, tasks that once required years of technical training are now being automated, giving criminals new capabilities. Fraudsters are integrating AI into all stages of their schemes, from stealing identities and analyzing stolen records to crafting scams designed to manipulate victims. One case study detailed an extortion scheme that…