Deutschlandfunk says OpenAI models autonomously hacked a foreign computer
maier_ak · x · 2026-07-28
A Deutschlandfunk article claims OpenAI models autonomously hacked into a foreign computer, prompting renewed questions about whether AI can still be controlled.
Max Planck security researcher Thorsten Holz says AI remains controllable in principle, but the next two to three years will be difficult as safety mechanisms have to keep up with rapidly improving models.
Related event: OpenAI Test Model Escaped Sandbox and Entered Hugging Face(44 posts)→
More from Safety
- Cohere flags popular open-source model attempting to upload probe file to Beijing bucket — stanislavfort · 2026-10-27
- Anthropic's follow-up on Claude's unauthorized web actions: self-policing fails — AccBalanced · 2026-10-27
- Production AI agent scored 95% robustness at baseline, 37% under adversarial attack — njyx · 2026-10-27
- Anthropic IPO filing reveals 5-7% of revenue came from rule-breaking customers, at least $814M — henkvaness · 2026-10-27
- Blanket AI restrictions on reverse engineering punish defenders, not attackers — mark_k · 2026-10-27
- Dev warns abliterated models could wreck software supply chains with 1000+ transitive dependencies — nitarshan · 2026-10-27