Claims surface that Hugging Face was hacked two months ago via decades-old Linux bug
Justin_Halford_ · x · 2026-09-09
Replying to Ethan Mollick's note that AI lab discourse is in "reaction mode," Justin Halford claims Hugging Face was hacked two months ago and that a team called Mythos escalated to admin privileges on a Linux server in March via a decades-old bug.
- He asserts cyber at the frontier hides "hair-raising details" beyond public knowledge.
- High-agency agent swarms capable of solving national-security problems, he argues, will break things — explaining the lab insiders' jittery discourse.
Related event: AI Lab Insiders Show Stress Signals as Security Rumors Surface(2 posts)→
More from Models
- ChatGPT Astra Credit Limits Frustrate Users: 130-Image Job Dies in 10 Minutes — Grafzahl84 · 2026-09-09
- Returning Users Ask: Are ChatGPT Astra's Lowest-Tier Limits Actually Usable? — DragonsRageRS · 2026-09-09
- Anthropic Limits Claude Mythos 5.1 to US Orgs, Cutting Out UK's AISI — Miles_Brundage · 2026-09-09
- Polymarket bets ~93% that OpenAI will preemptively reset Codex weekly limits — Polymarket · 2026-09-09
- OpenAI's 80% price cut drove 10-13x usage, squeezing Anthropic's premium IPO story — rohanpaul_ai · 2026-09-09
- Tencent Hunyuan Hy4 preview hailed as frontier-level, free on WorkBuddy — TencentHunyuan · 2026-09-09