Everything Sandbox Desk has reported about OpenAI: 5 stories, newest first. Part of our AI coverage.
Britain’s AI Security Institute will again test powerful models from OpenAI and Anthropic before public release, defying a recent White House directive that prioritized US reviews.
Experts criticize OpenAI's publication of solutions to 372 complex problems, citing concerns over human comprehension and scientific integrity.
Investors face challenges comparing AI leaders because Anthropic reports gross sales via cloud partners while OpenAI records only its net share.
The partnership focuses on training and evaluating autonomous agents for complex professional workflows, aiming to advance computer use capabilities.
OpenAI has flagged coordinated influence operations by Iran and Russia as its highest-rated threats, marking the first time it has used this severity classification.