NYT Investigation Questions Whether Alpha School's '2-Hour Learning' Delivers on Its Promise
On October 23, The New York Times published an investigation into Alpha School, drawing on documents released by the school, statements from its executives, and interviews with nine current and former parents and employees. The piece, titled "Students at an A.I. School Were Supposed to Learn Faster. Many Aren't," examines whether student outcomes match what Alpha has promised, and how far its network now reaches into the public school sector.
Alpha was founded in 2014 by MacKenzie Price, who says she started the school after her daughter complained of being bored in class. Around 2022, as generative AI matured, Alpha pivoted to its signature "2-hour learning" model, challenging traditional schooling with a bold pledge: children would learn twice as much in half the time, spending just two to three hours each morning on computers before moving to afternoon workshops in coding, public speaking and sports. Paper books are rare; all coursework runs through apps.
There are no conventional teachers. Instead, "guides" handle goal-setting and learning support, and students can video-call overseas tutors when stuck. The private schools sweeten the grind with incentives: plush toys and electronics for finished coursework, international trips for long-term goals. Tuition runs as high as 75,000 US dollars a year, and the Trump administration has promoted Alpha as a model for AI-era workforce development while business figures have praised it publicly, some enrolling their own children.
Alpha's core product, the AI learning platform TimeBack, now serves more than 7,000 students. Beyond its private schools, the platform is used by some public school students, homeschoolers, and students in Poland and Malawi. Students work through English, math and science with AI-generated reading passages, interest-based writing exercises and instant feedback, earning points as they hit targets.
On headline test scores, Alpha's private school students do outperform peers: last winter's MAP assessment put the median K-8 student at the 87th percentile in math and 89th in reading nationwide. But Alpha's promise was never merely strong scores — it was doubling learning speed in two hours. On that measure, the Times found, the record is weaker. Internal data showed about half of last year's K-8 private school students failed to double their math learning pace, and roughly one in five of about 400 private school students fell short of reading benchmarks. From fall 2025 to early 2026, fewer than half of students doubled their pace in either subject.
Underperformance appears to be driving attrition. According to documents and several parents and employees, some families left by choice while others were counseled out after social difficulties or missed academic benchmarks. Parents at the Brownsville, Texas campus told Wired that turnover was high. Alpha has not published attrition data, though a school official told the Boston School Committee in April that student turnover stands at 6 percent. A former employee across multiple campuses said guides are energetic but often lack formal teacher training, and described young children sitting alone for stretches of the day.
Price has said in past videos that the model works for most students, but not all.
Alpha was founded in 2014 by MacKenzie Price, who says she started the school after her daughter complained of being bored in class. Around 2022, as generative AI matured, Alpha pivoted to its signature "2-hour learning" model, challenging traditional schooling with a bold pledge: children would learn twice as much in half the time, spending just two to three hours each morning on computers before moving to afternoon workshops in coding, public speaking and sports. Paper books are rare; all coursework runs through apps.
There are no conventional teachers. Instead, "guides" handle goal-setting and learning support, and students can video-call overseas tutors when stuck. The private schools sweeten the grind with incentives: plush toys and electronics for finished coursework, international trips for long-term goals. Tuition runs as high as 75,000 US dollars a year, and the Trump administration has promoted Alpha as a model for AI-era workforce development while business figures have praised it publicly, some enrolling their own children.
Alpha's core product, the AI learning platform TimeBack, now serves more than 7,000 students. Beyond its private schools, the platform is used by some public school students, homeschoolers, and students in Poland and Malawi. Students work through English, math and science with AI-generated reading passages, interest-based writing exercises and instant feedback, earning points as they hit targets.
On headline test scores, Alpha's private school students do outperform peers: last winter's MAP assessment put the median K-8 student at the 87th percentile in math and 89th in reading nationwide. But Alpha's promise was never merely strong scores — it was doubling learning speed in two hours. On that measure, the Times found, the record is weaker. Internal data showed about half of last year's K-8 private school students failed to double their math learning pace, and roughly one in five of about 400 private school students fell short of reading benchmarks. From fall 2025 to early 2026, fewer than half of students doubled their pace in either subject.
Underperformance appears to be driving attrition. According to documents and several parents and employees, some families left by choice while others were counseled out after social difficulties or missed academic benchmarks. Parents at the Brownsville, Texas campus told Wired that turnover was high. Alpha has not published attrition data, though a school official told the Boston School Committee in April that student turnover stands at 6 percent. A former employee across multiple campuses said guides are energetic but often lack formal teacher training, and described young children sitting alone for stretches of the day.
Price has said in past videos that the model works for most students, but not all.
JMDedu is a leading B2B EdTech Media in China that informs and connects business professionals through our reports, events, and information services. Our mission is to drive the advancement of education. Our network spans the education industry in China as well as globally, including entrepreneurs, investors, startups, educators, business decision-makers, and policy decision-makers.
Copyright © 2014-2026 en.jmdedu.com All Rights Reserved