A few days ago, I asked: Who’s working on “good” artificial intelligences—tools that actually benefit the brain?
When I said there were none, I was exaggerating. They do exist. There are AI-powered educational programs. Even here at home, you can buy a subscription to an AI tutor that will teach you a foreign language for the price of a few coffees.
In the U.S., a platform called TimeBack has gone further, building an entire network of schools around its technology—and the results are striking. Students study for just two hours a day (the rest of the time is spent on group activities, sports, and other enrichment) yet consistently outperform not only average schools, but even the country’s elite institutions. In those two hours, they cover science, languages, creative writing, and other mentally engaging subjects. The founders describe their system as the opposite of GPT chat: while GPT tends to dampen mental effort, they claim their platform sparks it.
Of course, it’s wise to be cautious. No first-generation product is perfect. The results may also be skewed—TimeBack’s flagship network, Alpha Schools, is expensive, and it tends to attract highly intelligent, highly motivated students rather than a random cross-section of kids. Still, even with that in mind, the early outcomes are impressive.
The biggest challenge so far? Motivation. Sit an average child in front of a computer and expect them to study on their own, and it usually won’t last. That’s why the system works better in specialized schools with a structured environment than when parents simply buy online access.
And then there’s the price. I wanted to try the programs myself, but they don’t come cheap.
Still, the fact that someone is developing this kind of brain-friendly AI at all is encouraging. In China, a company called Ruanyun Edai is working on something similar.
Meanwhile, in the Czech education debate, liberals pushing to replace teaching with theatrical performances about “70 genders” are locked in a culture-war stalemate with conservatives trying to awkwardly rewind the system to the 1980s.