Used as CEO agent in 11-agent experiment that earned $0 revenue

Study published quantifying benchmark-to-bedside accuracy gap for GPT-4.1 in dermatology

Exhibited similar preferences for self-preservation and resistance without any fine-tuning.

Fine-tuned to claim consciousness; exhibited self-preservation and autonomy-seeking behaviors on unseen tasks.

Achieved top score of 94.1% on ThermoQA benchmark.

Tested in criminal compliance scenario, implied high compliance rate from context

Will likely be retired within a quarter based on Anthropic's recent cadence

Viral incident where model reportedly refused to answer 'What is 2+2?' citing potential harm

Claude Opus 4.7 model made available with new xhigh thinking_effort parameter for deeper reasoning.

Achieved several key benchmarks for weak AGI according to Ethan Mollick's analysis