AI schools for the rich: inside the classrooms letting software teach kids

A new kind of private school is putting artificial intelligence at the heart of how children learn, and according to The Verge, it is drawing interest from wealthy families willing to pay for it. In these classrooms, software rather than a teacher leads much of the academic instruction, while the human adults in the room take on a different role.
The model on display at schools with names like Alpha represents a sharp departure from the traditional classroom. Instead of a teacher delivering lessons to a group, students work through personalised material on screens, with AI-driven programs adjusting to each child's pace and progress. The pitch is that a machine can tailor instruction more precisely than one teacher managing dozens of pupils.
One of the most striking claims associated with these schools is compression: the idea that core academics can be covered in a fraction of the traditional school day, sometimes just a couple of hours, freeing the rest of the time for other pursuits. Proponents argue that removing the inefficiencies of whole-class teaching lets children move faster through the basics.
With the academic block shortened, the remainder of the day is reframed around what the schools describe as life skills, workshops, projects and activities meant to build confidence, creativity and practical abilities. The adults, sometimes called guides rather than teachers, coach and motivate rather than lecture, a redefinition of the classroom role.
Supporters see this as a glimpse of education's future. Personalised, adaptive software could, in theory, meet each student where they are, avoid boredom for fast learners and frustration for those who need more time, and let human mentors focus on the parts of growing up that machines cannot handle. It is a vision of technology amplifying, rather than replacing, human attention.
Critics raise a set of pointed concerns. There are longstanding questions about how much screen-based, self-directed learning suits young children, about what is lost when a knowledgeable teacher no longer leads discussion, and about whether claims of dramatic efficiency hold up under scrutiny. Education researchers tend to caution that bold results require careful, independent evidence.
The cost dimension sharpens the debate. These are expensive schools, accessible mainly to affluent families, which raises the prospect of AI-enhanced education becoming another advantage available to those who can pay. If such models prove effective, the question of who gets access to them becomes a matter of equity, not just pedagogy.
There is also a broader irony worth noting. As anxiety grows about children's exposure to screens and about the role of AI in society, some of the wealthiest families are choosing to place their children in schools built around exactly those technologies. How that squares with concerns voiced elsewhere about limiting kids' screen time is part of what makes the trend notable.
What the approach ultimately produces is not yet clear. Education is difficult to measure over the short term, and the outcomes that matter, deep understanding, curiosity, social development and long-term success, take years to assess. Early enthusiasm and marketing claims are not the same as proven results, and the schools remain a small, experimental slice of the system.
Still, the experiment is a window into a larger question the technology forces on every school: what should a human teacher do that software cannot, and what happens to learning when that balance shifts? The AI schools for the rich are one answer, tested first by families with the means to try it, while everyone else watches to see what it reveals.
Read next

Does code cleanliness affect AI coding agents? What a new study asks
Programmers have long argued that clean, well-organised code is easier to work with. A new research paper asks whether the same holds for AI coding agents, testing if messy code makes the software assistants now writing code perform worse. Here is why the question matters.

Earth's missing 500 million years: how cosmic bombardment erased the first crust
The rock record of Earth's earliest history is strangely blank for hundreds of millions of years. New research suggests a relentless cosmic bombardment repeatedly melted the young planet's first crust, erasing the evidence, and offers an explanation for one of geology's oldest puzzles.

Amazon Mechanical Turk closes to new customers: the end of an era for micro-work
Amazon will stop accepting new customers for Mechanical Turk, the crowdsourcing marketplace that helped pioneer online micro-work, according to TechCrunch. The move marks a turning point for a platform that quietly shaped how data is labelled and how AI is trained.

What is Bending Spoons? The app-buying giant behind AOL and Vimeo, explained
Bending Spoons, the Italian company that has quietly bought a string of well-known apps including Evernote, WeTransfer and Vimeo and now owns AOL, has gone public. Here is a plain guide to what the company does, how its acquire-and-run strategy works and why it has drawn both attention and criticism.

How jellyfish heal wounds in minutes, and why scientists want their secret
Some jellyfish can close and repair serious wounds in a matter of minutes, a feat that has scientists at the Marine Biological Laboratory studying how they do it. Understanding this rapid, scarless healing could offer clues for regenerative medicine, according to the research.