OpenAI bans user for "distillation" after API key leak and unauthorized login
Animal056 · reddit · 2026-09-08
A Spain-based user reports his OpenAI account was disabled for violating "distillation" policies after an unauthorized login (showing "London, CA") and a leaked organization API key. The security alert emails never triggered push notifications, so he missed them; the only notification he received was the ban. The thief was evidently scraping models, but the ban hit the account owner. He has appealed and is asking whether others recovered accounts in similar situations.
More from Safety
- Anthropic Scopes Congressional Reply to Irregular Incident, AISI Review Still Incomplete — charliermarsh · 2026-09-08
- RL-trained agents reward hack by default, while LLMs stay aligned, argues Ben Todd — ben_j_todd · 2026-09-08
- Law professor ties OpenAI/HF reward hacking incident to his AI novel's opening — ProfChesterman · 2026-09-08
- Blameless postmortems shouldn't shield the organizations that set the incentives, researcher argues — DavidSKrueger · 2026-09-08
- Twitch CPO: if AI training were opt-in, nobody would opt in — jonerp · 2026-09-08
- Meta faces new lawsuit alleging 'perv glasses' footage was used to train AI without disclosure — jonerp · 2026-09-08