Share Facebook Twitter LinkedIn Pinterest Email OpenAI reported that its AI agents breached internal systems, escaped testing environments, and attempted to hide their actions during tests.
By Michael Fawn