Asked when an AI could build a billion-dollar business on its own: AGI has already arrived.

Added to korrents 6 Sept 2026 · How quotes work · Something wrong? Tell us

Where this collides

• Cannot both be true

  AGI worth the name would combine human adaptiveness with machine reliability, not reproduce human performance including human failures.

  Huang, asked when an AI could build a billion-dollar business on its own, answers that AGI has already arrived. Marcus says AGI worth the name would combine human adaptiveness with machine reliability rather than reproduce human performance on tests. By that definition the thing Huang says has arrived has not been built.

• Cannot both be true

  Passing ARC-AGI does not amount to achieving AGI: o3 still fails on some very easy tasks, indicating fundamental differences from human intelligence.

  Huang, asked when an AI could build a billion-dollar business alone, answers that AGI has already arrived. Chollet says that even a system passing his own AGI benchmark has not reached AGI, because it still fails tasks a person finds very easy. They are not describing the same threshold, and only one of them thinks it has been crossed.

• In tension with

  There is a 25% chance of AGI by 2027, a 50% chance by 2034, and a 75% chance by 2045

  Alexander puts a quarter of the probability mass on AGI arriving by 2027 and half by 2034. Huang says it is already here. A forecast and a declaration of arrival are different kinds of statement, which is why this is a tension rather than a flat contradiction — but they cannot both be good descriptions of where 2026 sits.

Shared subject first, then similar wording. They may agree or disagree.

• A country outside the AI supply chain should just buy the index — which works only in the world where AI ends up commoditised rather than concentrated. AIAlex Imas4 Jun 2026AGI
• A human being is not an AGI: we lack a huge amount of knowledge and rely on continual learning instead, so continual learning is what superintelligence should mean. ISIlya Sutskever25 Nov 2025AGI
• A language model is no substitute for a well-specified conventional algorithm, so it cannot simply be dropped into a complex problem and trusted. GMGary Marcus7 Jun 2025AGI
• A poor country should prioritise owning a piece of AI over retraining its workers, but it should not bet everything on that. PTPhil Trammell4 Jun 2026AGI
• A strong track record in a technical field is only weak evidence that someone has good strategic judgment about AGI. NNNeel Nanda22 Mar 2025AGI

On the map

Open the map on its own page