Amit Sahai: AI is now producing new mathematical ideas, humans must keep up
In a guest post on Terence Tao's blog, cryptographer Amit Sahai argues AI systems are already generating original mathematical ideas and calls for a large expansion of human mathematicians to understand them.
What to know
- A guest essay by cryptographer Amit Sahai, posted on Terence Tao's blog, claims AI systems are already producing original mathematical ideas rather than just fast calculations.
- Sahai calls for a large, well-funded global expansion of mathematically trained researchers to keep pace with and understand AI-generated mathematical breakthroughs.
- He frames the stakes beyond academia, citing a hypothetical AI-proposed fusion power plant design as an example of why society needs humans able to understand such advances.
- The post has spread across Hacker News, Lobsters, and Mastodon bot reposts with modest engagement but little substantive public debate captured in available coverage.
Amit Sahai Guest essay authorTerence Tao Host mathematician, blog publisher
How it unfolded 2 developments, newest first · click a bar or a number to jump articlesposts
-
2
Story resurfaces on HN front page and spreads to Mastodon
A second Hacker News submission of the same essay reaches the front page, and automated Mastodon bots relaying Hacker News front-page stories repost the link.
“The AI systems I have worked with are already producing beautiful new ideas.”
— Amit Sahai, Guest essay author · source -
Without a massive cultural shift that leads to humans all, synchronously, choosing inefficiency over commercial growth, I just don't see this happening -- and as much as I'd like to engineer this cultural shift, I don't see how to. > For our community to give up the work of understanding would be a profound abdication of our responsibility to…
2 more of the top 3 · 63 posts in this stretch
-
> Before approving construction, I would want communities of humans to understand why the design works and what justifies confidence in its safety. I would hope that we all would.Until very recently, I pored over every single line of code Claude generated with razor sharp scrutiny. I would usually catch issues with every response. I'm catching…
-
R
Intriguing 🤔: “We’re Gonna Need A Lot More Mathematicians”, Amit Sahai via Terence Tao ( https:// terrytao.wordpress.com/2026/09 /24/were-gonna-need-a-lot-more-mathematicians/ ). Via HN: https:// news.ycombinator.com/item?id=4 9852717 # AI # LLMs # Mathematics # Math # Research # TheoremProving # Mathematicians # Vetting # ArtificialIntelligence
-
-
1
Post circulates on Hacker News and Lobsters
The essay is submitted and discussed on Hacker News and Lobsters, gaining modest engagement (dozens of points, single-digit to low-double-digit comments) as it spreads across tech aggregators.
-
background
Amit Sahai publishes guest essay on Terence Tao's blog — Sahai argues that AI systems are already producing original mathematical ideas beyond fast calculation, and calls on the mathematics community not to abandon research in the face of this pace, urging a large expansion of human mathematicians to understand and build on AI-generated ideas.
Also covered reported alongside — the timeline has no entry for these yet
-
first by HN Best, 1d ago · also HN Frontpage
What people are saying 21 voices from 2 sites · best of 63 · verbatim
- Yesterday
-
I feel like there’s a weird conflicts in how folks mental model of LLM development.It’s not the model I don’t trust, it’s myself. The model is wrong _all the time_ because - it’s easy to verify the code - it’s hard to verify that I knew what I was talking about when I prompted it.So the idea that you can broadly speaking take the human out of the…
-
I agree with the whole thing but as a tangent, I wish people would stop wishing for terawatts of power without saying in advance where they will put the entropy.
-
A metaphor I'm constantly drawn to is the transition from agrarian to urban societies following the Industrial Revolution. Somebody who somehow saw the Industrial Revolution coming from the perspective of somebody living in an agrarian society might have envisioned it leading to 'super farms.' And it did.But the biggest change wasn't what it did…
-
I find this discussion odd. 50% of people mistake this guest post for a post by Tao himself. The post is by Amit Sahai, professor for computer science and math at UCLA.Tao has been posting a flood of guest posts that stress inevitability and coping. Tao is fully invested in AI but needs to have the appearance of a broad discussion.Sahai is of…
-
I know many programmers for whom "developing domain understanding" is an abstract concept (if indeed, it is a concept for them at all). But in the age of big AI, I see more need for this, not less.At least this is my observation: when my colleagues have been wholesale chucking stuff over to Claude, they've then been confronted with classic…
-
The problem is not that we fear the AI will do maths we can do. At least that's not the problem. Whether the AI can write better code, devise better proofs, etc, I still want to do it. The thing that worries us (the generation trying to enter the job market now) is just our livelihood. Food, rent & accomodation while I do the work. "Who's gonna…
-
A lot of the conversations around AI in the past years have been “we will delegate almost all work to AI, but the most decisions will be left to humans.”I’m fairly confident that the opposite will be true. The most important decisions will be made by AI, and humans will only be left to guide decisions as a matter of taste. AI has the ability to be…
-
> I guess, the possibility of reclaiming human meaning and purpose.As you described very well, as humans we are mostly interested in solutions, not problems. You don't have to understand how a car works to make the most of it. Increasingly, you don't have to review every line of code to feel confident it is correct. But there is inherent value in…
-
> I have in mind something like a “deployable intellectual reserve”: communities of mathematically sophisticated people that humanity can call upon to help understand consequential AI-enabled breakthroughs.Can’t afford to be one of your credentialed reserves, Amit. ADHD screwed me over in early life or I’d have taken that dual masters in…
-
> The ideas and opinions presented here are entirely my own, but GPT 6 Astra was instrumental in helping me draft this note. Why did you really need GPT 6 Astra's help to write this note? Have we lost the ability to formulate a simple post like this without the AI's help?
-
No, for two reasons. First, we largely don't use wisdom or ethics today, efficiency is still the goal when humans are in charge. Companies and countries mostly operate on decisions to outcompete peers on relatively short horizons. I wish we made more decisions on ethics. Second, the current generation of AI is not sufficient for this, but attacks…
-
> To take on this responsibility, we may need to broaden our view of what a mathematician can contribute. I have in mind something like a “deployable intellectual reserve”: communities of mathematically sophisticated people that humanity can call upon to help understand consequential AI-enabled breakthroughs.> Our ability to understand difficult…
-
>Terence Tao is arguing that the human involvement in research is crucial, but doesn't convincingly justify why, in my opinion. He says that "human agency is a value of fundamental importance" and that we will need to build "thriving human communities that can understand [AI ideas] together" - not for the sake of correctness, which AI may surpass…
-
> even if AI is better able to understand and make the final call Maybe I'm misunderstanding, but are you claiming that a language model is better able to understand whether humanity as a whole should choose chase growth at the cost of developing as humans? To decide what the best values are (and not just the most "efficient" outcome)? If a human…
-
"We’re gonna need a lot more mathematicians."What the article really says is that we're going to need much smarter mathematicians. That is not possible for puny meat-brain humans. Humans are close to their ceiling. AIs are just getting started.In practice, we're probably going to hit that limit first in IC design. I once went to a talk by the…
-
There is a failure to understand that the process is the result. You don't study mathematics or computer science and information theory to produce commodities. You study them to transform your mind. The output of an LLM is useless without a human mind to comprehend it. We can have Super Intelligence, but if humans are incapable of comprehending…
-
A little meta but I really appreciate Terry's blog featuring a lot of guest posts recently. I'm not really a math person but following this debate(?) has been fascinating.
-
Like most of us I go back and forth between sheer optimism and fear for the future that AI may usher in. Recently I started vibe coding a fun video game with my ten year old. The experience is different than my work because it’s been such a joy to basically have a personal genie in a bottle help me make some personal art with a loved one…
- Sep 25
-
> This blog post was initially written in a different file format and converted using AI The juxtaposition of this next to the bold AI claims is odd. Did they not know how to copy-paste text? I would have hoped the AI would make the reference interactive links or something. In any case the thought experiment is interesting. We have a lot of…
-
> [This is a guest post by Amit Sahai. This blog post was initially written in a different file format and converted using AI. — T.]
-
> For our community to give up the work of understanding would be a profound abdication of our responsibility to humanity...to build a future in which humans can understand and contribute to the discoveries that will change our world. A future with meaningful human agency. This applies to pretty much every field where AI can be or could ever be…