Key Takeaways

- Alpha School offers 2 hours of AI tutoring daily plus project workshops, charging up to $75,000 annually
- Studies show AI homework help improves scores but drops exam performance by up to 24%
- The school plans two dozen new locations for fall 2025, targeting tech and finance hubs
Alpha School charges families up to $75,000 a year for two hours of daily AI tutoring followed by project workshops. The Austin-based school, now 12 years old, is expanding aggressively into wealth centers like Palo Alto, Malibu, and Manhattan. Parents in finance and tech are signing up, betting that traditional prep schools cannot prepare their children for an AI-transformed economy.
The Wall Street Journal reports that these schools replace teachers with "guides" or "coaches," use adaptive AI platforms that adjust lessons in real time, and promise personalized curricula that conventional classrooms cannot match. Every on-site learning guide at Alpha earns six figures. Billionaire Bill Ackman is reportedly among the school's fans.
What does the AI tutoring model look like?
Students spend two hours with an AI tutor each morning. The platform monitors engagement and adjusts difficulty, pacing, and content on the fly. Afternoons shift to project-based workshops led by human coaches. Alpha also sells its competency-based curriculum and homeschooling software to families who want the method without the campus.
San Francisco VC Shaun Johnson plans to enroll his son in Alpha's kindergarten. "We recognize that education is likely broken the way it is and there's going to be entrepreneurs that try to fix it," he told the Journal. He emphasized that AI-driven personalization, not the technology itself, drove his choice.
Alpha added eight locations in 2025. Nearly two dozen more are planned for fall, targeting markets where high-net-worth families cluster. New York families skew toward finance; Bay Area families come from tech.
Do AI tutors actually improve learning?
Two recent studies complicate the pitch. A Chinese study of over 26,000 students found that AI-assisted homework was faster and scored higher, but exam performance dropped by up to 24 percent. About 81 percent of long-term users simply outsourced their thinking to the AI. A UC Berkeley study reached similar conclusions.
The problem is not AI assistance itself. It is that students use AI as a replacement for thinking rather than a tool that enhances it. Traditional schools have no clear answer. They ban ChatGPT or ignore it, but neither approach teaches students to use AI productively.
Alpha claims to solve this by embedding AI into the learning process deliberately. The AI tutors, they argue, are designed to teach, not to do the work for students. Whether that design actually prevents cognitive outsourcing remains unproven.
The wealth divide in AI education
At $75,000 a year, this model is accessible only to wealthy families. The irony is sharp: AI tutoring is arguably the biggest equalizer for learning access in years. Anyone with internet can now access patient, adaptive, around-the-clock instruction for free or near-free. ChatGPT, Claude, and other models already function as personal tutors.
But using AI well requires skills that schools would need to teach first: knowing when to lean on AI and when to struggle through a problem yourself, evaluating AI output for accuracy, and maintaining the metacognition that long-term AI use seems to erode. The $75,000 schools claim to teach these skills. Public schools, stretched thin, mostly do not.
This creates a two-tier system. Wealthy families pay for structured AI education; everyone else gets the same tools but no guidance on how to use them. In San Francisco, where even six-figure salaries barely cover housing, OpenAI alone reportedly created 75 multimillionaires last fall. The wealth gap is widening, and education may widen it further.
Why traditional schools are stuck
Conventional schools face a structural problem. They are built around standardized curricula, age-based cohorts, and teacher-to-class ratios. AI enables one-to-one instruction, adaptive pacing, and mastery-based progression. These models are fundamentally incompatible.
Most schools have responded by treating AI as a cheating tool to police. Some have experimented with AI-assisted grading or tutoring pilots. But few have rethought the core model, which is what Alpha and similar schools claim to do.
The question is whether Alpha's approach actually works. The school is 12 years old, which means early cohorts should be graduating. Longitudinal data on college admissions, career outcomes, and cognitive development would settle the debate. That data does not appear to be public.
Logicity's Take
For AI product teams, the education market reveals a familiar pattern: incumbents cannot integrate AI because their core processes resist it, while startups build from scratch. Alpha School is essentially a vertical AI SaaS company wrapped in a school. The bet is that AI tutoring plus human coaching outperforms traditional instruction. That bet may be right, but the evidence is thin. Teams building AI tutoring products should note the engagement-tracking approach and the emphasis on adaptive content, but also the risk: if the Chinese study's 24% exam drop replicates, the whole model collapses. Watch for outcome data.
Frequently Asked Questions
How much does Alpha School cost?
Tuition runs up to $75,000 per year. Every on-site learning guide earns a six-figure salary.
Where is Alpha School located?
Alpha School was founded in Austin, Texas. It added eight new locations in 2025 including San Francisco and New York, with nearly two dozen more planned for fall in places like Palo Alto and Malibu.
Does AI tutoring improve student performance?
Studies show mixed results. AI-assisted homework scores higher, but a Chinese study of 26,000 students found exam performance dropped by up to 24% among AI users, with 81% of long-term users outsourcing their thinking.
What is the daily schedule at AI private schools like Alpha?
Students spend two hours with AI tutoring each morning, followed by project-based workshops led by human coaches in the afternoon.
Who is enrolling in AI private schools?
Wealthy families from finance and tech industries. New York families tend to work in finance or run businesses; Bay Area families come from tech. Billionaire Bill Ackman is reportedly a fan.
Examines how AI tools fail when users don't know what to ask, parallel to AI tutoring's cognitive outsourcing problem
Need Help Implementing This?
Building AI tutoring or adaptive learning products? Logicity covers the tools, frameworks, and strategies that matter. Subscribe for weekly analysis on AI product development.
Source: The Decoder / Matthias Bastian
Manaal Khan
Tech & Innovation Writer
Produced with AI assistance and reviewed by the Logicity editorial team. Learn more in our Editorial Policy.
Related Articles
More in AI & Machine Learning
Bezos AI Lab Gets $10B: What Project Prometheus Means
Jeff Bezos is closing a $10 billion funding round for Project Prometheus, an AI lab focused on physics-based AI for manufacturing and engineering. With a $38 billion valuation and backing from JPMorgan and BlackRock, this signals a major shift in enterprise AI investment toward industrial applications.

Kimi K2.6 Open-Weight AI: 300 Agents at a Fraction of the Cost
Moonshot AI's Kimi K2.6 matches GPT-5.4 and Claude Opus 4.6 on coding benchmarks while running 300 parallel agents. For businesses locked into expensive API contracts, this open-weight model could slash AI infrastructure costs while delivering enterprise-grade automation.




