Germany and Europe need more political urgency on AI
Reflections after a conversation with a member of the German parliament on AI safety, European models, chips, electricity and sustained political attention.
Tracking down traces of rogue agent swarms
Agent Swarms: What do they know? Do they know things? Let's find out!
substackHow I Use Codex To Automate Parts Of My Research Workflow
How I use Codex, scoped access, research triage, cheap experiments, loops, and durable project knowledge to reduce repeated research friction.
substackWe are watching intelligence become a gated asset class
Frontier AI isn't closing so much as moving behind gates. Notes on tiered access, concentrated infrastructure, and what it means when powerful machine intelligence starts to function like privately controlled infrastructure.
Monthly Papers — January 2026
Learning to do research better: lessons from Nanda, Perez, and Eisner on exploration, truth-seeking, and prioritisation.
Creativity
Reflections on creativity and how I nurture it in my life. Thoughts on the value of ideas and the importance of acting on them.
Monthly Papers — November 2025
This month’s reading: concrete AI futures, how fragile “normal life” might be under powerful tech, and a couple of papers on training/evaluating LLM agents.
Monthly Papers — October 2025
A brief summary of this month's reading. Papers read and interesting findings.