Ex-DeepMind Researcher: AI Governance Should Target Compute, Software, and Networks

Andrea Michi, ex-DeepMind and now CTO of AI safety startup depthfirst, argues that maintaining control over AI should be pursued through levers at the compute, software, and network layers.

2026-09-18 ~ 2026-09-18 · 2 related posts

• Ex-DeepMind researcher on AI control: compute, software and networks are the intervention points — andreamichi · 2026-09-18
• Ex-DeepMind researcher's depthfirst essay: nuclear-age lessons for staying in control of AI — andreamichi · 2026-09-18