Brian Chau is a mathematician and policy analyst who writes at fromthenew.world, where he explores rationality, epistemology, AI alignment, and civilizational trajectories. His work critiques institutional consensus and examines how elite narratives evolve — or collapse — under the weight of social pressure, follow Brian on Twitter
We talk about the danger of elite overreach, the misuse of AI policy, the failures of both left and right narratives, and how to build something more coherent. Topics are outlined in the timestamps below.
Listen on Apple Podcasts, Spotify, or wherever you get your podcasts.
Support my work here and follow me on Twitter here
Timestamps
00:00 – Intro
02:12 – Export controls, global alignment, and the US/China divide
04:24 – The politics behind AI regulation and compute limits
06:36 – Bipartisan handling of AI across Trump and Biden
08:48 – Bridging partisan divides on technological progress
11:00 – Antitrust, incumbents, and regulatory capture
13:12 – Global AI competition and the Chinese model
15:24 – Risk aversion, family obligations, and talent bottlenecks
17:36 – Leaving AI orgs and reevaluating the mission
19:48 – Perplexity AI vs DeepSeek: the Tiananmen test
22:00 – Open-source LLMs and political resilience
24:12 – Consent, compliance, and values in AI behavior
26:24 – Group disparities and admissions philosophy
28:36 – Ethical tradeoffs in policy modeling
30:48 – Surveillance as a control system
33:00 – Misunderstanding distant cultures and narrative bias
35:12 – Bottom-up development vs Western paternalism
37:24 – Economic diversity beyond Silicon Valley
39:36 – Final reflections and historical insight
Share this post