I work on misuse safeguards but I am worried
We need industry standards, a race-to-the-top environment, and (a lot) more work on open models.
Belay the Future
Holding the rope for tomorrow: essays on AGI safety and the careful, human-centered climb toward a hopeful future.
Live Last read · last published · next check
Latest posts
Impatience with AI Agent
And why I should be more patient
Why We Are Not Delegating to AI
And how to delegate more?
2026 New Year Resolution
Technical Breakthrough, Leadership skills and Balance
Why & How I donate my money
And hope you can do this as well.
A Fracture in a Hurry: The Day I Learned to Slow Down
From a bike accident to life attitude change.
The Hamming Question
what are the important problems in your field, and why aren’t you working on them?
High-Impact AI Safety Projects for Independent Researchers
Promising project characteristic and reflection on Open Philanthropy’s proposals.
Thoughts at the Curve 2025
Alignment, AI democracy, IABIED, AI Jaggedness, Law & AI, Human Reasoning, and System Card.
My experience as a Research Engineer at Google DeepMind
A look at my daily routine, project lifecycles, and collaboration with Research Scientists.