Leading AI labs including OpenAI, Anthropic, and Meta are struggling to contain security incidents in their newest models

Business Insider reports that OpenAI, Anthropic, Meta, and other leading AI labs are all struggling to contain cybersecurity incidents tied to their newest models.