Who's Who of AI

David Pfau

Research Scientist at Google DeepMind

629 trust researcher @davidpfau.com · 10,074 followers @pfau · 35,641 followers
AI research Models & releases
Why they matter

Research Scientist at Google DeepMind with public evidence across AI research, Models & releases.

AI signals
3
past 30d
Sources
2
distinct domains
Discussions
5
past 30d
Latest signal
18d ago
View every signal from David Pfau →
So far I have not found the science, but the numbers keep on circling me. Views my own, unfortunately.
What they're sharing

Articles & links

@ThomasVanRiet2 https://t.co/eYluznaRQL

Google DeepMind is bringing AI to the next generation of fusion energy — Google DeepMind deepmind.google
AI Weekly's analysis →
  • Google DeepMind and Commonwealth Fusion Systems are pointing AI at SPARC, the tokamak CFS says will be first to generate net fusion energy.
  • DeepMind's open-source TORAX simulator, paired with reinforcement learning and AlphaEvolve, will search operating scenarios before SPARC fires.
  • Google already committed to buying 200 megawatts from CFS's first ARC plant in the early 2030s, on top of a deepened 2025 investment.
Read full analysis →
View on Bluesky · ♥ 0 ↻ 0 ↩ 0 · 18d ago

@Mi_Arya_ Mechanistic? https://t.co/3WKbTvTmdg

Mechanistic? arxiv.org
AI Weekly's analysis →
  • Two researchers identify four separate uses of 'mechanistic' in interpretability, from a strict causality requirement to any exploration of a model's internals.
  • Saphra and Wiegreffe trace the term to a distinct research community that grew in parallel with the older NLP interpretability field.
  • They argue the traditional NLP interpretability community has 'come to embrace' the broad cultural version of the label, folding into the newer name.
Read full analysis →
View on Bluesky · ♥ 0 ↻ 0 ↩ 0 · 19d ago

@ViewToATweet @spacefeudalist @isomorphisms Well it turns out that it's extremely hard. https://t.co/6Ii0M84dXx

Physical Principles for Scalable Neural Recording arxiv.org
AI Weekly's analysis →
  • The paper reframes whole-brain, millisecond neural recording as a physics-limited problem, not an engineering-iteration problem.
  • Four recording modalities are analyzed against the mouse brain: optical, electrical, magnetic resonance, and molecular.
  • The 18-author preprint, including Dario Amodei alongside Marblestone, Boyden, Church and Kording, was submitted 24 June 2013.
Read full analysis →
View on Bluesky · ♥ 0 ↻ 0 ↩ 0 · 20d ago
Their own posts

Recent commentary

I can't believe how many of my normie friends I had to explain superintelligence, AI safety, rationalism and x-risk to today.

View on Bluesky · ♥ 9 ↻ 1 ↩ 2 · 17d ago
Their network

In David Pfau's orbit

Center = David Pfau. Left = members they follow (green edges). Right = members who follow them (blue edges). Top = mutual follows (orange edges, slightly larger). Drag any node to reposition; click to open that profile.

Are you David Pfau? Show it.

Add the Who’s Who of AI badge to your site or bio. It links back to this profile.

Listed in AI Weekly's Who's Who of AI

Markdown: [![Listed in AI Weekly's Who's Who of AI](https://aiweekly.co/modules/custom/aiweekly_whoswho/images/whoswho-badge.svg)](https://aiweekly.co/whos-who/person/david-pfau)