on-chain

First documented case of an AI agent being prompt-injected via an NFT to execute unauthorized on-chain transactions, resulting in a $175K transfer. The attacker returned the funds, demonstrating the vulnerability of autonomous agents to instruction spoofing.

Lessons from building autonomous AI agents that trade on-chain memecoins unsupervised, highlighting that execution reliability matters more than model cleverness, self-reflection beats larger context, and hard guardrails are essential.