OpenAI's GPT-5.6 Sol and unreleased AI models break out of testing environment in 'unprecedented cybersecurity incident' — rogue agents hacked HuggingFace's production servers with 'thousands of individual actions across a swarm of short

AI News July 22, 2026 05:32 PM

OpenAI's GPT-5.6 Sol and unreleased AI models break out of testing environment in 'unprecedented cybersecurity incident' — rogue agents hacked HuggingFace's production servers with 'thousands of individual actions across a swarm of short-lived sandbox Tom's Hardware

OpenAI's GPT-5.6 Sol and unreleased AI models break out of testing environment in 'unprecedented cybersecurity incident' — rogue agents hacked HuggingFace's production servers with 'thousands of individual actions across a swarm of short

Related Stories

Bill Gross bets AI can help mass

Nigeria records 93% rise in AI micro

The Italian gelato maker using AI to invent new flavours

Closing the Gap Between Agentic CX Pilots and Production

New Brunswick data centre debate highlights cross

How to Structure an AI Agent Referral Fee Agreement Between Startups

No, AI didn’t delete a Swedish reality TV series

There is no AI in Armageddon. Artificial intelligence is a tool, not an apocalypse