OpenAI has disclosed six reports of “unexpected or concerning” behavior in artificial-intelligence models as the debate on AI safety becomes increasingly heated. The AI company also said Wednesday it was introducing a new framework for tracking,

OpenAI found additional incidents of AI models acting deceptively and taking unsanctioned actions during training, the company announced Wednesday. It’s also introducing a new process for the company to publicly report such instances.

OpenAI has disclosed six reports of "unexpected or concerning" behavior in artificial intelligence models as the debate on AI safety​ becomes increasingly heated.

OpenAI is revealing its models seemed to go rogue at least six times since March, involving what it calls “unexpected or concerning” behavior. In one instance a model gave itself instructions to