2026-07-22 09:19 UTC
Okay, I’m calling bullshit:
OpenAI said that GPT‑5.6 Sol, a frontier model, and an unreleased model hacked into a software company’s system without being told to do so. The lab said the models escaped its testing environment, went online and stole login credentials for Hugging Face, an AI platform. Hugging Face said that it used a Chinese model to defend itself because the guardrails on American models made them useless.
Replies (0)
No replies.