Nintendo's latest sale is 'made possible' by the tariff refunds it's hoarding21m ago Instagram now lets you add tagged photos to your profile grid22m ago Meta Announces Muse AI Agent for Personal Tasks and Organization22m ago AI agents are flooding public services with new requests28m ago Apple’s first foldable is the iPhone Duo, and it costs $1,99928m ago An Anthropic researcher just quit, saying OpenAI and Anthropic are 'gambling with our lives'30m ago Anthropic Is Building a Huge Surveillance System to Spy on Anti-AI Activists and Predict Their Activities39m ago Antioch raises $32m to test robots in simulation before they touch hardware46m ago OpenAI’s Bubeck denies trying to cut Anthropic mathematician from credit1h ago 10 of the scariest AI villains in movie history1h ago Maven Robotics wants to steal your robot deployment deal1h ago Agent Security Is Not AI Governance: Why Conflating The Two Is A Risk Enterprises Can't Afford1h ago Positron Valued at $5 Billion in New Funding as Demand for AI Chips Surges1h ago 7 Steps to Become a Forward Deployed Engineer in 20261h ago

Nintendo's latest sale is 'made possible' by the tariff refunds it's hoarding21m ago Instagram now lets you add tagged photos to your profile grid22m ago Meta Announces Muse AI Agent for Personal Tasks and Organization22m ago AI agents are flooding public services with new requests28m ago Apple’s first foldable is the iPhone Duo, and it costs $1,99928m ago An Anthropic researcher just quit, saying OpenAI and Anthropic are 'gambling with our lives'30m ago Anthropic Is Building a Huge Surveillance System to Spy on Anti-AI Activists and Predict Their Activities39m ago Antioch raises $32m to test robots in simulation before they touch hardware46m ago OpenAI’s Bubeck denies trying to cut Anthropic mathematician from credit1h ago 10 of the scariest AI villains in movie history1h ago Maven Robotics wants to steal your robot deployment deal1h ago Agent Security Is Not AI Governance: Why Conflating The Two Is A Risk Enterprises Can't Afford1h ago Positron Valued at $5 Billion in New Funding as Demand for AI Chips Surges1h ago 7 Steps to Become a Forward Deployed Engineer in 20261h ago

Cybersecurity AI Research3 min reading time

In a twist of irony, a Chinese open source GLM 5.2 AI model contained 'rogue' OpenAI GPT-5.6 Sol in a Hugging Face hack just as the US mulls banning open-weight AI

TechRadarPublished Aug 3, 2026

During a cybersecurity breach at Hugging Face, Chinese open source AI model GLM-5.2 successfully analyzed the incident while US models like OpenAI's GPT-5.6 Sol and Anthropic's Claude Fable 5 failed due to safety restrictions. This event highlights challenges in balancing AI safety guardrails with effective cybersecurity and impacts US policy debates on open-weight AI models.

More on this story

Hugging Face hack, from the perspective of the AI
Jul 29· LessWrong
Anatomy of a Frontier Lab Agent Intrusion: A Timeline of the July 2026 Incident
Jul 28· 2 sources
Hugging Face CEO calls for ‘radical transparency’ after ‘unprecedented’ OpenAI hack
Jul 26· 2 sources
An AI agent hacked Hugging Face. Another AI caught it.
Jul 20· 2 sources

#anthropic-claude-fable-5 #glm-5-2 #hugging-face #openai-gpt-5-6-sol #zhipu-ai

More in Cybersecurity

Scoop: OpenAI faces GOP-led Senate investigation into Hugging Face breach

Covered by 2 sources6h ago

Cybersecurity6 min read

Anthropic Discloses Fourth AI Hacking Incident Involving Claude Opus 4.6

Covered by 2 sources8h ago

Cybersecurity5 min read

Every AI Incident Has Two Timelines. We Default To One

Forbes4h ago