Currently doing AI Safety Research at FAR.AI.
Previously finished my PhD, reverse-engineering neural networks to understand how they think, so we can try to stop them from thinking bad things.
- Doing Research on technical alignment at FAR.AI - see my scholar page.
- Procrastinating with Side projects:
- AI-Safety-Papers — a living reading-list with concise notes.
Website | Twitter/X @afspies | LinkedIn | ✉️ alex [at] afspies (dot) com