OpenAI Calls for Global AI Standards and Alignment Research Amid 'Concerning Model Behavior'

OpenAI advocates for international cooperation to establish global technical standards for frontier AI models, focusing on benefit-risk management and alignment research. This call comes as the company reports new instances of 'concerning model behavior' and intense debate about AI safety.