Real Alpha
Try Pro free

MTS · 2 min · Nov 4, 2026

AGI and superintelligence are distinct concepts

Richard Ngo · Researcher, OpenAI

Richard Ngo says people blurred AGI and superintelligence partly because they expected recursive self-improvement to close the gap.

In this episode

Full analysis
  1. 12:37Running a company exposes AI’s gaps in reliability4 min
  2. 19:19Alignment research should clarify what good behavior means3 min
  3. 22:31AI safety needs groups willing to challenge powerful labs5 min
  4. 5:00Rapid self-improvement timelines outrun the evidence5 min
  5. 1:39AI safety's deference to labs reflects fear of conflict3 min
  6. 10:30AGI and superintelligence are distinct concepts2 min
Up next · Latent SpaceBundled market data fixed connector failuresPro pick 2:37
“it's become apparent that AGI is very different from superintelligence.”
Richard Ngoat 11:37 · MTS
11:37

Takeaways

  1. Earlier arguments often blurred human-level AI and superintelligence.
  2. He says the conflation has made past arguments confusing.
  3. He expects “superintelligence” to need more precise subcategories.

About this segment

Richard Ngo says people blurred AGI and superintelligence partly because they expected recursive self-improvement to close the gap. He expects the broad label superintelligence to break into more precise concepts.

Richard Ngo

Richard Ngo is an independent AI researcher; former OpenAI governance and DeepMind safety researcher.

Researches multi-agent intelligence after working on AI alignment at DeepMind and governance at OpenAI.

  • Theo Jaffee, Host · Co-founder & host, MTS

    One of the first members of a16z's new media team; earlier hosted the Emergent Ventures-backed Theo Jaffee Podcast.

  • Sophia Dew, Host · Host, MTS

From “AI Cannot Reliably Run a Company for Six Months | Richard Ngo”, Nov 4, 2026. Watch the full episode

Transcript

Transcribed by machine (Whisper) from the episode’s audio, so a word may be off here and there.

  1. 10:35

    So you wrote a, you know, short form post about this on LessWrong. Daniel Cocotagelo responded and you had like a back and forth thread. And Daniel basically said, well, like your claim doesn't really mean anything because things in the future will be so weird such that to the average person or to, I guess, any forecaster, It will be indistinguishable from, you know, the effects of true superintelligence, even if the machines that we're dealing with are not exactly true superintelligence.

  2. 11:09

    Yeah, my sense is that. So one way I put it is that people used to use the word, the term AGI as a catchall for, you know, incredibly sophisticated AI. And they didn't really distinguish much between AGI and superintelligence, partly due to arguments about recursive self-improvement. Like once, like probably there's not a huge gap between, say, a human level AI and a superintelligent AI was the argument.

  3. 11:37

    And I think over the last five years, let's say, it's become apparent that AGI is very different from superintelligence. And that we, yeah, like people who conflated those two terms have mostly ended up being confused. I think they'd say, well, you know, when we said AGI, we kind of meant superintelligence. But like, you need to get the terminology, like precision with this kind of stuff is important. So my sense is that the term AGI has now been like, we have much more clarity on what that means and what that looks like. And now the term superintelligence has replaced that as the thing that is sort of this, you know, catch all term for incredibly powerful AI. We don't know exactly what we mean by superintelligence, but we're sort of bundling together a whole bunch of different concepts.

  4. 12:26

    uh and my sense is that over the next few years we're going to have to unbundle superintelligence as a term and we're going to see like yeah like even the concept of superintelligence is going to need to get quite spiky is another way of putting it right so you