Skip to content
Quotes.

Ilya Sutskever quote, Nov 25, 2025

Co-founder, Safe Superintelligence · Nov 25, 2025 · Interview, Dwarkesh Podcast: We're moving from the age of scaling to the age of research

The thing which I think is the most fundamental is that these models somehow just generalize dramatically worse than people.

Primary source

Words found on the source page on Sep 25, 2026.

More from Ilya Sutskever

Quote
All the pre-trained models are pretty much the same because they pre-train on the same data.Nov 25, 2025· Interview
But now that compute is big, compute is now very big, in some sense we are back to the age of research.Nov 25, 2025· Interview
I think that’s been guiding me a fair bit, thinking from multiple angles and looking for almost beauty, beauty and simplicity.Nov 25, 2025· Interview
If you think about the term “AGI”, especially in the context of pre-training, you will realize that a human being is not an AGI.Nov 25, 2025· Interview
Number three, I think it would be really materially helpful if the power of the most powerful superintelligence was somehow capped because it would address a lot of these concerns.Nov 25, 2025· Interview
One consequence of the age of scaling is that scaling sucked out all the air in the room. Because scaling sucked out all the air in the room, everyone started to do the same thing.Nov 25, 2025· Interview
So for research, you definitely need some amount of compute, but it’s far from obvious that you need the absolutely largest amount of compute ever for research.Nov 25, 2025· Interview
So it’s back to the age of research again, just with big computers.Nov 25, 2025· Interview

On the same topics

Quote
An AI with more real layers can actually be “smarter” in the conventional sense of the term.Scott AlexanderSep 24, 2026· Post
For 100 steps in a row, the AI can shoot delicate subtle ideas from layer to layer at the speed of light. Then there’s one step where it has to encode them into twenty-six glyphs invented by Phoenician turquoise miners in 1800 BC. Then it has to re-encode the Phoenician glyphs into delicate subtle lightspeed ideas before it can do anything else.Scott AlexanderSep 24, 2026· Post
The science of reading these thoughts is a subfield of AI interpretability, which is still in its infancy.Scott AlexanderSep 24, 2026· Post
This trains the AIs to be focused on task success, which naturally risks including things like reward-hacking, cheating, and single-minded pursuit of stated goals at the expense of ethical injunctions.Scott AlexanderSep 23, 2026· Post
It is important to not confuse massive steps in inference-time scaling, a dynamic which should be fairly predictable, with being the outputs of RSI, which is highly uncertain.Nathan LambertSep 19, 2026· Post
And actually the models today are far beyond what was possible even six months ago.Noam BrownSep 17, 2026· Interview
About this quote
Said by
Ilya Sutskever, Co-founder, Safe Superintelligence
Where
Dwarkesh Podcast: We're moving from the age of scaling to the age of research
Topics
Research, AGI
Company
Safe Superintelligence
Source
dwarkesh.com/p/ilya-sutskever-2

A quote is one moment, not the whole argument. Read it in context at the source.

Sources: the primary source linked on each quote, checked weekly. A quote is one moment; read the source. Logos via logo.dev; trademarks belong to their owners.

New quotes by email

Fridays, only in weeks with new on-the-record quotes from data and AI leaders.

Double opt-in. Unsubscribe any time.