Why Tech Alone Won't Save Our Schools: Lessons From Sal Khan

Sal Khan warns that EdTech needs human teachers to succeed, as studies reveal virtual schools struggle without intense human oversight and tutoring support.

Tuesday, July 28, 2026

Key Takeaways

  • Khan Academy founder Sal Khan warns that educational technology fails on its own. It requires structured teaching and human interaction to work.
  • Research backs this up. A national CREDO study found that students in virtual charter schools lost the equivalent of an entire school year (180 days) of math learning. Similarly, a University of Kentucky study showed that high schoolers who transitioned to virtual charter schools had lower graduation rates and higher dropout rates than their peers in traditional public schools.
  • But technology can succeed when paired with personal instruction. Newark Public Schools combined Khan Academy with a $400,000 high-impact tutoring grant. With this hybrid approach, the district achieved state math test score gains triple the New Jersey average.

Educational technology shows impressive results in specialized environments, but its creators warn that software is not a magic cure for learning loss. Khan Academy founder Sal Khan recently urged educators and parents to credit teachers, not just technology, for recent school successes. Technology-enabled classrooms can thrive, but their achievements depend on heavy human support and structural advantages that are difficult to replicate.

What Happened

In a recent essay published by Khan Academy, founder Sal Khan described the results of two schools built around his mastery-learning model. At the private Khan Lab School in Silicon Valley, high school students routinely master core subjects by the eighth grade, and the recent graduating class achieved a median SAT score of 1520. Meanwhile, the tuition-free Khan World School, an online charter school run with Arizona State University, saw its inaugural class of ninth graders make up to 4.6 times the typical annual progress in math. According to a Freakonomics podcast interview, University of Chicago economist Steve Levitt initially doubted these figures. He said he had never seen such rapid growth in 25 years of education research, but he confirmed the results after analyzing the raw data.

Khan warns against attributing these outcomes solely to software. The Khan Lab School relies on a tuition of over $30,000 to fund high teacher salaries, and it attracts heavily invested families alongside high-profile guest speakers. Even the free Khan World School model relies on highly selective, self-motivated students. These students spend most of their time in live Socratic discussions, intensive reading, and Oxford-style advisor check-ins. Khan warns that expecting technology alone to transform education without these heavy investments in human interaction will lead to failure.

The Bigger Picture

Broader academic research supports Khan’s cautious outlook. Virtual charter schools generally struggle, even if specific virtual schools do well under ideal conditions. A study by the University of Kentucky found that students who transition to virtual charter high schools face lower graduation rates, higher dropout rates, and lower college enrollment than peers in traditional schools. Data compiled by the Human-AI K-12 Evidence Project shows that virtual charter students lost the equivalent of an entire school year of progress in mathematics. In contrast, physical charter schools, like those in Oklahoma, often show strong academic gains. This indicates that physical, school-level environments remain critical.

When technology succeeds in public schools, it is almost always paired with intense human intervention. In Newark, New Jersey, public schools achieved math gains nearly triple the state average, as reported by The Learning Standard. But this success required a 70 percent student engagement rate achieved through structured school programs. It also relied on a $400,000 state grant that paid for students to meet with the same tutor three times a week. Even with these resources, Newark's overall math proficiency still lags behind state averages. The recovery process remains slow and difficult.

What This Means for Families

For parents and educators, the lesson is clear. Technology is a useful tool to free up time, but it cannot replace human teaching. While mastery-based learning improves equity and helps diverse learners when implemented well, it also carries risks. Research published in an education conference analysis warns that self-paced learning can lead to severe procrastination and minimal effort if students lack external accountability.

Families choosing online or self-paced options must ensure the program includes regular teacher check-ins, Socratic discussions, and peer interaction. Software delivers personalized practice, but humans must provide the motivation, guidance, and structure to keep students on track.

What You Can Do

  • Ask about the exact ratio of live, teacher-led instruction to independent screen time when evaluating virtual or blended school models.
  • Set weekly progress goals if your child uses self-paced mastery software to prevent procrastination.
  • Ask your local school district if they offer targeted tutoring programs paired with interactive software, rather than letting students use digital tools in isolation.
Share:
Why Tech Alone Won't Save Our Schools: Lessons From Sal Khan | The Learning Standard