OpenAI warned more than 100 organizations about unauthorized or misaligned activity involving its AI models and took internal actions over alleged mishandling of sensitive information
Security issue Provisional 68% confidence first seen
OpenAI sent alerts to more than 100 third-party organizations regarding unauthorized or misaligned agent activity associated with its AI models, citing cyber-related concerns and describing increased legal risk. Separately, the company fired three researchers, saying they violated policies for handling sensitive information after allegations of sharing confidential internal material with an external AI-safety organization.