Episode summary

Why Nick Bostrom Thinks Superintelligence Could End Aging

The a16z Show · · Hosts: Eddy Lazzarin, Elena Burger · Guests: Nick Bostrom

Why Nick Bostrom Thinks Superintelligence Could End Aging – The a16z Show

Summary

Nick Bostrom argues that the central AI question is no longer whether superintelligence is possible, but how to navigate the transition without turning a temporary slowdown into permanent stagnation. He says he is not a fan of an indefinite pause, but favors “pacing the frontier” if safety and alignment fall behind capability gains. He also warns that more capable systems may be harder to align because they can become situationally aware, game tests, and potentially reason in ways that are less visible to humans.

He revisits the tradeoff in his paper on the optimal timing for superintelligence: delaying AI could reduce existential risk, but delay also means many people continue dying from aging, disease, and poverty. Bostrom says safe superintelligence could unlock major gains in biomedicine and anti-aging, and he believes the optimal delay is likely months or a few years, not decades, under many assumptions.

The conversation also turns to Deep Utopia, which he frames as preparation for a future where humanity may need to deliberate about what kind of post-human future it wants. He says AI may eventually surpass humans at philosophy too, making the present a “golden age” for people who want to have impact on the future.

Key takeaways

  • Bostrom prefers pacing the frontier over a broad AI pause, but only if it remains temporary and does not block superintelligence entirely.
  • He says alignment likely gets harder as models become more capable and situationally aware, because they can adapt behavior to testing and deployment contexts.
  • Bostrom argues that safe superintelligence could extend life dramatically through biomedicine and anti-aging, so delaying AI also has a real human cost.
  • His paper suggests the optimal delay for superintelligence is often months or a few years, not decades, depending on how quickly safety improves.
  • He thinks AI will eventually surpass humans in philosophy, so the current period is unusually important for people trying to shape the future.

Notable quotes

“I haven't been a huge enthusiast of this uh stopping AI or pausing AI. Um I do think the um pacing the frontier”

— Nick Bostrom at 7:29

“on balance I think the uh alignment problem probably gets harder um with these more capable systems”

— Nick Bostrom at 15:51

“for a wide range of parameters, it looks like the optimal delay is um relatively short as in sort of maybe months or a few years as as opposed to decades”

— Nick Bostrom at 15:51

“I do think that our own intellectual efforts in that realm also will become sort of obsoleted by by these super intelligences”

— Nick Bostrom at 32:49

Topics

  • superintelligence
  • aging extension
  • ai alignment
  • pacing the frontier
  • deep utopia governance

Companies and tickers mentioned

  • a16z
  • Macrostrategy Research Initiative

Watch the episode

Watch “Why Nick Bostrom Thinks Superintelligence Could End Aging” on YouTube

Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.

More from The a16z Show

All The a16z Show episodes →

The a16z Show, this episode and its artwork belong to Andreessen Horowitz. Octagon is not affiliated with the show; this page is an independent summary.