News
Reasoning models struggle to control their chains of thought, and that’s good
3+ hour, 20+ min ago (340+ words) 5 de mar'o de 2026 Why a limitation of frontier models is reassuring for AI safety. Chain of thought refers to the intermediate reasoning steps an AI agent generates while solving a task. These reasoning traces can provide useful safety signals, helping…...
Scaling social science research
2+ week, 11+ hour ago (365+ words) Scaling social science research'OpenAI Scaling social science research A new tool to help researchers turn qualitative data into numbers they can analyze. A core part of our work at OpenAI is enabling scientists to move faster and solve harder problems....
Advancing independent research on AI alignment
2+ week, 10+ hour ago (582+ words) Advancing independent research on AI alignment'OpenAI Advancing independent research on AI alignment We're committing $7.5M to The Alignment Project to fund independent research developing mitigations to safety and security risks from misaligned AI. Today, we're announcing a $7.5 million grant to The…...