AI AGENTS FIND $4.6M IN BLOCKCHAIN SMART CONTRACT EXPLOITS

Summary from the TLDR Dev newsletter

Anthropic researchers evaluated AI agents' ability to exploit smart contracts on a newly created benchmark, SCONE-bench. The agents, including Claude Opus 4.5 and GPT-5, successfully exploited vulnerabilities in smart contracts, collectively amassing $4.6 million in simulated stolen funds from contracts exploited after March. In a separate simulation, the AI agents uncovered two novel zero-day vulnerabilities in recently deployed contracts, showing the technical feasibility of profitable, real-world autonomous exploitation.