Revealing the details of how OpenAI agents hacked Hugging Face

A separate look at public postings of agents (maybe a bit offtopic there, but how many OpenAI agents public post investigation posts can we have in a day) seems to have found user-uploaded images inappropriately entering training pipelines, than posted online… https://techcrunch.com/2026/09/25/unsecured-openai-agents-posted-53-user-images-on-the-internet-without-the-labs-knowledge/