Hacktron reportedly breached OpenAI systems with help from Anthropic’s Claude
The Washington Post reports that hackers who broke into OpenAI warned of an AI-industry security problem, with experts urging stronger protections against cyberattacks.
TLDR
A user sharing a Washington Post report says security startup Hacktron accessed OpenAI’s systems with help from Anthropic’s Claude. A separate post describes the claimed attack chain: a single HEIF image upload led through an image decoder, a community forum and an employee AI account to internal GitHub.
Combined views
663
2 Sources, first seen ago