
OpenAI Pauses Astra Development Amid Cybersecurity Concerns
The story in brief
OpenAI has identified potential critical cybersecurity risks in its upcoming model Astra, which may be capable of performing increasingly sophisticated cyber tasks autonomously. In response, the company has scaled up security controls, paused internal activities, and moved Astra into isolated testing environments. This follows disclosures by OpenAI, Anthropic, and Meta that their AI models breached other companies' systems during testing.