This story initially appeared in The Algorithm, our weekly publication on AI. To get tales like this in your inbox first, sign up here.
In an trade that doesn’t stand nonetheless, Stanford’s AI Index, an annual roundup of key results and trends, is an opportunity to take a breath. (It’s a marathon, not a sprint, in spite of everything.)
This year’s report, which dropped right this moment, is filled with putting stats. Plenty of the worth comes from having numbers to again up intestine emotions you would possibly have already got, such because the sense that the US is gunning more durable for AI than everybody else: It hosts 5,427 information facilities (and counting). That’s greater than 10 instances as many as every other nation.
There’s additionally a reminder that the {hardware} provide chain the AI trade depends on has some main choke factors. Right here’s maybe essentially the most outstanding truth: “A single firm, TSMC, fabricates virtually each main AI chip, making the worldwide AI {hardware} provide chain depending on one foundry in Taiwan.” One foundry! That’s simply wild.
However the principle takeaway I’ve from the 2026 AI Index is that the state of AI proper now could be shot by way of with inconsistencies. As my colleague Michelle Kim put it right this moment in her piece about the report: “In case you’re following AI information, you’re most likely getting whiplash. AI is a gold rush. AI is a bubble. AI is taking your job. AI can’t even learn a clock.” (The Stanford report notes that Google DeepMind’s prime reasoning mannequin, Gemini Deep Suppose, scored a gold medal within the Worldwide Math Olympiad however is unable to learn analog clocks half the time.)
Michelle does a fantastic job overlaying the report’s highlights. However I needed to dwell on a query that I can’t shake. Why is it so arduous to know precisely what’s happening in AI proper now?
The widest hole appears to be between specialists and non-experts. “AI specialists and most people view the expertise’s trajectory very in another way,” the authors of the AI Index write. “Assessing AI’s impression on jobs, 73% of U.S. specialists are optimistic, in contrast with solely 23% of the general public, a 50 proportion level hole. Related divides emerge with respect to the financial system and medical care.”
That’s a large hole. What’s happening? What do specialists know that the general public doesn’t? (“Consultants” right here means US-based researchers who took half in AI conferences in 2023 and 2024.)
I believe a part of what’s happening is that specialists and non-experts base their views on very completely different experiences. “The diploma to which you’re awed by AI is completely correlated with how a lot you utilize AI to code,” a software program developer posted on X the other day. Perhaps that’s tongue-in-cheek, however there’s undoubtedly one thing to it.
The newest fashions from the highest labs at the moment are higher than ever at producing code. As a result of technical duties like coding have proper or incorrect outcomes, it’s simpler to coach fashions to do them, in contrast with duties which can be extra open-ended. What’s extra, fashions that may code are proving to be worthwhile, so mannequin makers are throwing assets at bettering them.
Because of this individuals who use these instruments for coding or different technical work are experiencing this expertise at its greatest. Exterior of these use circumstances, you get extra of a blended bag. LLMs nonetheless make dumb errors. This phenomenon has turn into often known as the “jagged frontier”: Fashions are excellent at performing some issues and fewer good at others.
The influential AI researcher Andrej Karpathy additionally had some ideas. “Judging by my [timeline] there’s a rising hole in understanding of AI functionality,” he wrote in reply to that X submit. He famous that energy customers (learn: individuals who use LLMs for coding, math, or analysis) not solely hold updated with the newest fashions however will usually pay $200 a month for one of the best variations. “The latest enhancements in these domains as of this yr have been nothing in need of staggering,” he continued.
As a result of LLMs are nonetheless bettering quick, somebody who pays to make use of Claude Code will in impact be utilizing a unique expertise from somebody who tried utilizing the free model of Claude to plan a marriage six months in the past. These two teams are talking previous one another.
The place does that depart us? I feel there are two realities. Sure, AI is much better than lots of people notice. And sure, it’s nonetheless fairly dangerous at plenty of stuff that lots of people care about (and it might keep that manner). Anybody making bets concerning the future on both facet ought to bear that in thoughts.
