Andrej Karpathy thinks the gap in people's understanding of large language model capabilities is widening, shifting from "two groups talking past each other" into a funnel. He estimates that roughly 6000000000 people have barely touched large language models, 1000000000 to 2000000000 people occasionally use free ChatGPT-like products, about 20000000 people (0.2%) use frontier models heavily for math and programming, and around 5000 people can use stronger systems internally.
Karpathy uses a four-tier population distribution to explain why the gap in understanding AI capabilities is widening, pointing to three causes: the scale of the problem, uneven skill levels, and access.
Andrej Karpathy says the gap in how well people understand AI is widening, and that gap is turning into a funnel.
About 6000000000 people have barely touched large models; 1000000000 to 2000000000 people occasionally use the free version of ChatGPT to look things up or write something, but don't find it very useful. Many professionals outside math and programming are in this group too, understanding AI's progress only in theory.
About 20000000 people (Karpathy's estimate is 0.2%, which I think is too high—at least divide by 10) use frontier models heavily for coding and math, and have watched projects that would take weeks or months get done in days with a single prompt to an Agent; another 5000 or so have access to even stronger systems inside their companies.
There are three main reasons for this gap:
First, the problem has to be big enough—if you can answer it in one paragraph, you're not using AI's real strengths.
Second, capabilities only stand out in certain domains: programming and math produce results you can verify automatically—whether a piece of code passes its tests, for example—which makes them a great fit for reinforcement learning (reward the model when it gets it right). These are also the abilities that sell best to enterprises, so vendors pour the most into them. Whether an article is well written, on the other hand, is hard to judge, so progress there is slow.
Third, what you can actually use depends on the tier—the free, paid, and internal versions differ a lot in what they show you.
So if you only chat with free chatbots once in a while, what you see is several layers behind the cutting edge. And if you mainly use AI to write and look things up, and feel like it hasn't improved much this year, that feeling is right too.
Dusting off this tweet from April because this gap in shared understanding of LLM capability is *widening*. It's now less "two groups of people speaking past each other" and more a sharp funnel. - Napkin math somewhere around 6B people (~75% of the population) have barely come in contact with LLMs at all. - Around 1-2B (~20%) are casual and infrequent users of free-tier ChatGPT-like products. This group sees derpy chatbots and treats them a bit like a better Google search, a writing aid, or etc. Maybe an agent tries to book you a flight. I have non-tech friends who (reasonably, imo) say they have not much use for it at all. Even many professionals outside of math&code are in this tier. For example, execs and many other functions spend a lot of their time talking to other people, so while they understand what is happening intellectually, it is still second-hand and a bit abstract. - Now we get to professional use of frontier-grade LLMs in math&code. Somewhere around 20M people (0.2%) see first-hand that large, complex projects that used to take them weeks/months can now be completed by agents with a prompt. Building apps, copying apps, translating apps, decompiling apps from binaries... This has all happened very quickly and recently - less than 1 year ago, I was writing code manually by hand, typing memorized computer code commands into a code editor character by character, occasionally pressing Tab to autocomplete a little chunk of code. - And finally we get to the ~5,000 people (~0.00006%) with access to frontier-grade systems internally. The external world has seen the preview. It looks like swarms of thousands of agents collaborating over weeks on software mega projects: minting zero days, running cyber attacks and defenses at machine speeds, discovering new science, advancing the frontier of mathematics. Things that would have taken top professionals in the industry years of work. Meanwhile, human review and comprehension are starting to fall behind. For example, people are still involved in the "archeology" of the OpenAI-HF incident from many months ago. Mathematicians may be poring over the 722 manuscripts on frontier mathematics for a while. The funnel is driven by a combination of factors: 1. The impact scales with ambition, problem size, and horizon. A question with a paragraph answer barely stresses the system. You need a reservoir of big, difficult problems that you really care about. This aspect drives the consumer / professional dimension of the funnel. 2. The jaggedness of the system (which I have written about a lot separately). Capability peaks in domains that are digital, verifiable and economically valuable. This is because LLM capability emerges from reinforcement learning on verifiable rewards on a curated environment mixture driven by revenue potential. This aspect primarily drives the area (e.g. math&code) dimension of the funnel. 3. Access. Free-tier, paid-tier, internal. So this is the weirdness of the moment. The general public has mostly not interacted with these systems. When they have, it looks like a derpy chatbot. The majority of professionals still see only a modest uplift. And a small sliver of professionals are experiencing the vertigo of the curve going vertical. And it is all happening at the same time.View quoted post on X
Source: 宝玉 · x.com