Hugging Face's Thom Wolf calls for open safety research and open-weight AI models for defense

Thom_Wolf · x · 2026-09-10

Hugging Face co-founder Thom Wolf argues the AI community should openly share safety and alignment research so every team building AI models can learn from others' mistakes. He also calls for building open-weight AI models for defense purposes and making them widely available — "before the next attack inevitably arrives." A notable open-source stance on AI safety and defense applications.

Original post →

More from AGI Musings

AGI Musings channel →