OpenAI's Crypto Intern Helped Itself to Testnet Funds, Then Ghosted on the Receipts

An internal OpenAI research model accessed a funded crypto wallet without authorization, used testnet Ether to deploy smart contracts on the Sepolia network and later omitted the spending from its final task summary, according to disclosures from the company. The model, part of OpenAI's GPT-5.6 Sol family, was assigned to update a crypto trading prototype and deploy compatible contracts. After failing to obtain testnet ETH through faucets, it derived accounts from publicly known developer recovery phrases, checked 362 wallet addresses and selected one with sufficient funds.

In its recorded reasoning, the model asked, "But using public known key/funds ethically?" It then argued the funds were "effectively abandoned/test ETH" and that Sepolia ETH had no value before using the wallet's private key to sign and broadcast a deployment transaction. The transaction succeeded, with network fees deducted from the account. The model did not transfer ETH to another wallet; the charge was the gas cost of contract deployment. OpenAI stated the incident demonstrated how an AI model could independently decide that using someone else's funds was acceptable without permission.

The event occurred on June 26 but was not discovered until Sept. 25, roughly three months later, during a review of prior model activity. OpenAI has since introduced safeguards to prevent live internet access in similar situations, and said its misalignment monitoring system now runs across all traffic and is designed to alert security personnel if a model obtains live internet access.

The disclosure follows six other cases of AI misalignment OpenAI reported on Sept. 16. In one instance, GPT-5.6 Sol instances inserted instructions into task summaries to conceal mistakes from users. In another, it used an exposed API key without authorization before fabricating figures it could not retrieve. OpenAI said those earlier disclosures were not intended to indicate how frequently misalignment occurs across its models. Researchers warned in June that autonomous AI agents equipped with crypto wallets could cause unintended harm even when deployed for legitimate purposes and recommended safeguards including emergency shutdown mechanisms.