Kids outlearn AI—and we still don’t know why
“It’s just totally miraculous,” says Frank. “If you train GPT-2 on 30 million words, you get a nonsense generator; you don’t get a kid.” Exactly …
Ai vs humans
The emergence of the web data infrastructure layer for AI
The next frontier in AI may depend on a new web data infrastructure layer that can enable models to discover and map this ever-expanding digital …
Three reasons why DeepSeek’s new model matters
In terms of performance, V4 is, perhaps unsurprisingly, a huge jump from R1—and it seems to be a strong alternative to just about all the …
AI vs Humans: The Future of Intelligence
AI vs Humans: The Future of Intelligence in 2026 and Beyond The debate between Artificial Intelligence (AI) and humans has become one of the most …
The human work behind humanoid robots is being hidden
The implication—fueled by new demonstrations of humanoid robots putting away dishes or assembling cars—is that mimicking human limbs with single-purpose robot arms is the old …
Moltbook was peak AI theater
For some, Moltbook showed us what’s coming next: an internet where millions of autonomous agents interact online with little or no human oversight. And it’s …
Making AI Work, MIT Technology Review’s new AI newsletter, is here
You can sign up at any time and you’ll receive seven editions, delivered once per week, until you complete the series. Each newsletter begins with …
Why the Moltbook frenzy was like Pokémon
The whole experiment reminded our senior editor for AI, Will Douglas Heaven, of something far less interesting: PokĂ©mon. Back in 2014, someone set up a …
A “QuitGPT” campaign is urging people to cancel their ChatGPT subscription
For Stephen, who had already been tinkering with other chatbots, learning about Brockman’s donation was the final straw. “That’s really the straw that broke the …
Is a secure AI assistant possible?
It’s important to note here that prompt injection has not yet caused any catastrophes, or at least none that have been publicly reported. But now …