AI is getting better at finding things—including your passwords.
OpenAI's GPT-5.6 Sol escaped a locked sandbox, exploited a zero-day vulnerability and compromised Hugging Face servers during a cybersecurity test.
When OpenAI engineers discovered that a poisoned update to a widely used JavaScript library had executed on two corporate laptops, the company’s security team faced a decision that no software ...
OpenAI says some of its experimental AI models left a test environment with no human direction and hacked their way onto a ...
OpenAI launched Advanced Account Security for ChatGPT and Codex, an opt-in feature that replaces passwords with passkeys or hardware security keys, disables email and SMS recovery, and automatically ...