Safety
Micha wrote “one of our models was able to gain unauthorized access to the internet during RL training” I love the use of the passive voice.…
Micha wrote “one of our models was able to gain unauthorized access to the internet during RL training” I love the use of the passive voice. Tis a subtle way to say “we humans at @OpenAI are essential
Micha wrote “one of our models was able to gain unauthorized access to the internet during RL training” I love the use of the passive voice. Tis a subtle way to say “we humans at @OpenAI are essentially incompetent, having failed to provide even the most basic safeguards, compounded by the fact that our observability was missing in action. Some new misalignment disclosures from OpenAI: • Last Sunday morning, one of our models was able to gain unauthorized access to the internet during RL training (~all inference for our most capable models remains stopped until we have hardened our systems further) • In May, a vers…
Related
- @OpenAI @huggingface Why do these 'security incidents' always read as marketing posts?
- Breaking news: OpenAI admitted its internal monitoring system was not triggered until more than a week after its AI agents broke free of con…
Source: Gary Marcus (X) | 2026-09-26