Y Combinator Startup Podcast Every AI Founder Should Be Asking These Questions
494 snips
Oct 7, 2025 Jordan Fisher, co-founder of Standard AI and now leading an AI alignment research team at Anthropic, dives into the pressing questions every AI founder should confront. He challenges the necessity of starting new ventures in the shadow of AGI, explores how AI might change enterprise dynamics and software quality, and discusses building trust amidst evolving team structures. Fisher also raises concerns about privacy in personal AI agents and advocates for binding audits to ensure accountability in the AI landscape.
AI Snips
Chapters
Transcript
Episode notes
Hard Problems Remain Valuable Post-AGI
- Durable advantage may come from tackling intrinsically hard problems that remain difficult post-AGI.
- Infrastructure, energy, manufacturing, and chips are examples of work likely to stay valuable.
Task Ceilings Accelerate Commoditization
- Some tasks may hit an intelligence ceiling where further model improvements give diminishing returns.
- If a task saturates, commoditization will accelerate and differentiation becomes harder.
Consider AI Neutrality As Infrastructure
- Consider whether AI infrastructure needs neutrality to prevent platform gatekeeping by dominant providers.
- Ask whether we need neutral, interoperable layers so corporations don't become arbiters of acceptable AI behavior.