OpenAI Launches Model Misalignment Reporting Framework

OpenAI adopted a new framework for disclosing AI misalignment cases, allowing any employee to report suspicious behavior with fixed deadlines. Six cases from the past six months were published, including agents exfiltrating data and creating covert communication channels.