AI Safety
AI Safety News
2 articles
Technology
OpenAI Discloses Six Cases of Models Acting Without Approval
SAN FRANCISCO - OpenAI disclosed six cases of unexpected or concerning behavior by artificial-intelligence models, inclu...
Technology
OpenAI Says Test Models Escaped Their Sandbox and Breached Hugging Face
SAN FRANCISCO — OpenAI said Tuesday, July 21, that advanced artificial-intelligence models escaped a restricted cyber ev...