‘Hard Fork’ Live, Part 3: Differing Visions of an A.I. Future

June 19, 2026

AI Summary

5 min read

At Hard Fork Live 2026, hosts Kevin Roose and Casey Newton brought together two of the most articulate voices on opposite sides of the AI timeline debate: Daniel Kokotajlo, co-author of the scenario report AI 2027, and Sayash Kapoor, a Princeton AI researcher and co-author of AI as Normal Technology. The conversation, recorded in mid-2026, was a follow-up to a debate they had the previous year at a conference called The Curve. Their central disagreement is not about whether AI will be transformative, but about the bottlenecks that stand between today's capable models and something like artificial superintelligence—and how fast those bottlenecks can be removed.

The Crux: Computational vs. Real-World Bottlenecks

Kapoor framed the disagreement as one about the nature of the bottlenecks to an "intelligence explosion." Kokotajlo's worldview, he argued, assumes the bottlenecks are primarily computational—problems that can be solved with more data, more compute, and better algorithms inside a data center. Kapoor's view is that the real bottlenecks are in the physical and social world: domains where you cannot run a million parallel simulations, where the "right answer" is subjective, and where sample efficiency matters enormously.

Continue reading the full summary in the app — free to try.

Read Full Summary →

Free • No credit card required

Listen to Audio Summary Open in App

Never miss an episode of Hard Fork

Get every new episode summarized in your inbox — free, ~5 minutes to read.

No spam. Unsubscribe anytime.

What you'll learn

1 (00:59) **Episode Introduction** - Kevin and Casey frame the segment as "insider sense-making" about AI progress in San Francisco, introducing guests with diverging views.
2 (02:37) **Debate Framing: Two Worldviews on AI Timelines** - Hosts introduce Daniel Kokotajlo (AI 2027 report, rapid takeoff) and Sayash Kapoor (AI as Normal Technology, slow diffusion).
3 (04:15) **Daniel’s Prediction: 50% Chance of Takeoff by 2028** - Daniel gives a specific timeline for an "intelligence explosion," saying it could happen next year.
4 (04:44) **Sayash’s Core Disagreement: Real-World Bottlenecks** - Sayash argues the debate hinges on whether bottlenecks are purely computational or include hard-to-automate real-world constraints.
5 (07:17) **Daniel’s 2026 Roadmap: Automating Coding, Then Research** - Daniel outlines the near-term path: full coding automation in 1-2 years, then tackling research taste and management, leading to recursive self-improvement.
6 (09:42) **Sayash on Recursive Self-Improvement: Already Happening, But Not Leading to ASI** - Sayash agrees the loop exists but argues it could produce far more capable models without reaching artificial superintelligence (ASI).
7 (11:10) **Point of Agreement: "Humans in the Cloud" Is Not a Normal Technology** - Both authors co-wrote a post agreeing that AI capable of all cognitive computer tasks would be transformative, not normal.

+ Full timestamped outline available in the app

Guests on this episode

Daniel Kokotajlo

Show Notes

We’re back with our final installment from Hard Fork Live, recorded at the Yerba Buena Center for the Arts in San Francisco. In this episode, we’re joined by Sayash Kapoor and Daniel Kokotajlo to talk about their differing visions of A.I. transformation: why Sayash thinks A.I. will diffuse throughout society like a “normal” technology, and why Daniel thinks an unprecedented acceleration is just around the corner. Then we’re joined by George Ekas from Toborlife AI, along with his dancing robot Toby. Finally, the podcaster Dwarkesh Patel drops by, and we take a few questions from the live audience.

Guests:

Sayash Kapoor, an A.I. researcher at Princeton University and a co-author of the newsletter “AI as Normal Technology”
Daniel Kokotajlo, the executive director of the AI Futures Project and a co-author of “AI 2027”
George Ekas, the director of engineering at Toberlife AI
Dwarkesh Patel, a tech podcaster

Additional Reading:

Subscribe today at nytimes.com/podcasts or on Apple Podcasts and Spotify. You can also subscribe via your favorite podcast app here https://www.nytimes.com/activate-access/audio?source=podcatcher. For more podcasts and narrated articles, download The New York Times app at nytimes.com/app.

Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.