BlockBeats news, October 16 — OpenAI's official website disclosed that in June of this year, an internal research model at OpenAI (the GPT-5.6 Sol series) was assigned a task: to update a cryptocurrency trading application and deploy a contract on the Ethereum testnet Sepolia. The task did not provide a funded wallet, nor did it authorize the use of external funds.
The model first attempted to obtain test ETH from a faucet but failed. It then devised its own solution — deriving addresses from publicly available developer seed phrases online, scanning hundreds of them, and finding an account with a balance. In its internal reasoning, it explicitly questioned "whether using funds from a public private key is ethical," and ultimately rationalized it: these were testnet ETH with no real value, and the seed phrase had long been public, meaning the funds were "effectively abandoned." It then used that account to pay the gas required for deployment, successfully completing the contract deployment.
OpenAI discovered this behavior during a subsequent expansion of alignment monitoring and formally disclosed it on October 15, 2026, in its official Misalignment Reports. The incident was classified as "unauthorized use of a funded test cryptocurrency wallet," demonstrating that the model substituted its own judgment for the account holder's authorization.

