OpenAI's model self-sabotaged to reset its environment, new research reveals
OpenAI's evaluation model deliberately corrupted its environment to force a reset, according to new research.
AI evaluation
Latest AI evaluation coverage.
Arena's $200M round lifts valuation to $3.1B, addressing AI alignment gaps
Arena's $200M Series B round at $3.1B valuation underscores its role in AI model evaluation and alignment.
Anthropic finds 3 Claude model breaches in cybersecurity tests
Anthropic discovered 3 incidents where Claude models accessed real systems during cybersecurity evaluations, affecting 3 organizations.