Bulletin boards and shelves of worn paperbacks still line the walls of American classrooms, but virtually every surface now holds a charging station or a mounted screen. Students sit in rows, faces illuminated by the blue glow of their district-issued tablets. Teachers move among desks, but their primary interactions with students happen through “learning-management systems” where they monitor progress in real time.

In some sense, students today are getting a more “personalized” education than ever before—content is matched to their pace and learning style, which digital platforms measure more precisely than human teachers ever could. Adaptive software offers “personalized learning paths,” which adjust math problems or language lessons based on past performance. Yet many educators can’t shake the impression that technology has fundamentally warped the role of the teacher, and the quality and purpose of education. When students face the screen, the distinctive pedagogical qualities of the teacher fade into the background. Algorithms have recast teachers as consultants, facilitators, managers of classrooms that are no longer meaningfully their own. The above is not a depiction of computer class; it’s just class, in 2025.

In the last 30 years or so, America’s classrooms have undergone a quiet revolution. This didn’t happen overnight; nor did it result from a single policy decision. It emerged from a perfect storm of good intentions, corporate interests, and perverse incentives that took root in the 1990s. Progressive state governments, concerned about preparing students for an increasingly digital world, partnered with NGOs and technology companies to develop curricula around “digital citizenship.” In those days, there was a clear line between “computer class,” with its focus on typing and basic software skills, and the rest of the curriculum. Students would spend an hour learning simple programs via screen, then return to traditional instruction for core subjects. That boundary began to blur in the 2000s as “ed-tech” companies proliferated, promising to change how students learn everything from multiplication to Shakespeare. Federal grants incentivized schools to adopt these digital tools, while a growing consulting industry convinced administrators that staying competitive meant staying digital. By the late 2010s, what had started as an admirable goal—ensuring that students could navigate the emerging Internet age—became a Trojan horse through which technology companies captured the infrastructure of American education.

I. The Rise of the Digital Classroom

Today, the idea of a non-digital classroom seems as antiquated as a one-room schoolhouse. Thousands of schools across the country—primary and secondary, public and private, religious and secular—have surrendered education to screens and software. Students navigate their academic lives through a web of learning management systems, adaptive platforms, and digital assessment tools. Their reading is tracked word by word, their mathematical skills analyzed click by click. Their movements are recorded into vast databases that promise to optimize learning through algorithmic precision. The line between “digital learning” and simply “learning” has disappeared entirely. The result is an altogether different education system, which serves the interests of ed-tech companies far better than it serves students, teachers, schools, or society at large.

In a 2015 report on the impact of computers in schools, the Organization for Economic Cooperation and Development (OECD) notes that “students who use computers very frequently at school do a lot worse in most learning outcomes.” In a comprehensive meta-analysis of 126 studies, the Poverty Action Lab similarly concludes that “initiatives that expand access to computers . . . do not improve K-12 grades and test scores, [while] online courses lower student academic achievement compared to in-person courses.” In the U.S., the largest study of its kind among K-12 students finds that “even small daily amounts (30min) of use of digital devices in classrooms are negatively related to scores on a reading-comprehension test.” Another meta-study shows that even investing in air conditioning benefits student learning more than investing in laptops or tablets. The list of studies goes on, but the empirical consensus has been clear for some time. The ed-tech revolution has corroded the foundations of learning. As we’ll show in what follows, it has failed on its own terms, and not for lack of resources or support. On the contrary, despite the billions of dollars in investment and the hundreds of public-private partnerships, ed tech has failed because its proponents fundamentally misunderstand what education is, and the role education is designed to play in the formation of the young.

Some education reformers have begun to protest the ubiquity of screens in the lives of kids, noting the ways in which they dominate children’s social, educational, and leisure time. Last year, the Jewish Parents Forum (JPF), an organization of parents and educators committed to transmitting Jewish and Zionist values, led a revolt against smartphones in schools. (Note that the JPF is a project of Tikvah, which also publishes Mosaic.) Partnering with Jonathan Haidt, whose bestselling 2024 book, The Anxious Generation, documents the harms that screens inflict on children, the JPF convinced dozens of Jewish day schools to ban smartphones from their campuses. Early reports of the results have been striking—more concentration within the classroom, more vibrancy beyond the classroom, and more eye contact in general. This phone-free policy has been adopted by some secular schools too; Suzanna Kruger, a high-school biology teacher from Seaside, Oregon, noted in a recent report in the Free Press that, when her students had phones in class, “they just seemed so overcome with ennui. And now that they don’t have them, they seem so young again—so fresh and excited.”

Yet according to Caroline Bryk, the JPF’s executive director, phone bans in schools are vital but insufficient to reverse these deleterious trends. Bryk puts it bluntly: “We spent a year convincing schools to ban phones, but our kids still spend most of their day in front of a screen. And they have no choice: it’s literally part of the curriculum. If [Haidt] is right about phones, isn’t he also right about screens in general?”

Thousands of parents across the country have had to contend with this question. Many intuit that spending hours staring at screens is harmful to children, or at least not as beneficial as time spent with friends or playing outside or reading physical books. Parents thus often try to control their children’s screen time at home, but such efforts seem futile so long as most classroom hours are spent in front of an iPad. “It feels like everyone understands that screens are bad for kids, but teachers still don’t want to take the tablets away,” one worried parent explained. Even at most phone-free schools, there is no consistent principle applied to screen use. And more often than not, teachers have no say in the matter. Administrators impose “ed tech” in classrooms, weakening the authority that teachers wield over their own lesson plans. Parents blame teachers, teachers blame administrators, administrators blame principals, and principals and district bureaucrats blame the state.

In public schools, the dominance of ed tech is enshrined in yearslong partnerships among legislatures, school boards, progressive nonprofits, and ed-tech companies. Private schools, however, are in a better position to reform from within. And Jewish schools, in particular, could provide a blueprint for others to follow. Two years ago, when Bryk and Haidt started their campaign to ban phones in Jewish day schools, the situation seemed just as irreversible. But they found a way to build a new consensus. Ed tech represents another destructive trend decades in the making—and another occasion for Jewish schools to show the way forward.

II. How Pedagogy Replaced Education

The story of ed tech begins in the 1990s, but the philosophy behind ed tech began much earlier. In 1958, sensing dangerous trends in the American classroom, Hannah Arendt published “The Crisis in Education” in Partisan Review. In the essay, Arendt lamented the rise of an ever-expanding caste of school bureaucrats, who were increasingly tasked with supervising teachers and dictating the terms of curricula. Arendt also deplored the emergence of a “checklist” approach to education, in which the experience of group study was reduced to a set of sterile “skills” to be imparted mechanically. And she lambasted the widespread obsession with a mechanized approach to “pedagogy,” a discipline created to replace the charisma and personality of educators with uniform tools meant to keep students superficially engaged.

Arendt doesn’t identify particular targets for criticism in her essay, but it seems to be aimed at progressive reformers like the immensely influential John Dewey, who sought to develop a unified system of teaching in the United States. Among other things, Dewey pushed for a form of study in which teachers would operate as mediators of educational experiences, rather than authoritative guides. He died in 1952, just a few years before Arendt published “The Crisis in Education.” By that point, his life’s work was bearing fruit, particularly his goal of attaining an ideal democracy through the severe imposition of equality in educational methods. Arendt noted the widespread feeling in America that “Johnny can’t read”—that things had in fact gotten worse since Dewey’s push for a utopian, experimental, one-size-fits-all approach to learning. As she wrote, “for the sake of certain theories, good or bad, all the rules of sound human reason were thrust aside.”

The innovations promoted by progressive reformers promised to empower teachers and produce measurable results. Yet they robbed teachers of authority and classrooms of character. Once trusted to teach reading and writing using the approach best suited to their particular students, teachers now had to follow a standard curriculum that mandated specific texts and methods. Once in charge of evaluating their own students, teachers were now subject to external evaluations in the form of standardized tests. Once free to develop their own voices as educators, teachers were forced to attend “pedagogy workshops” and adopt prefabricated “learning techniques.” For Arendt, these transformations disfigured the art of teaching, fragmenting the activity of education into discrete parts to be optimized and evaluated. In the name of “transparency” and “efficiency,” bureaucrats had undermined the purpose of education as well as the teacher’s role as an example and leader. Each classroom was no longer a world of its own, carefully crafted by an individual whose charisma or kindness or humor lent itself to a distinctive form of study. In the standardized classroom, too much personality generated suspicion, and conformism invited praise.

The results were not auspicious. In 1983, the National Commission on Excellence in Education issued a report documenting sharp and steady declines in SAT scores and other measures since the 1960s. In 1998, the Hoover Institution revisited and updated the report, noting that another “lost generation” had passed through high school since the initial report’s publication. Along with many other findings, the Hoover report notes the many millions of Americans who graduated high school without being able to read at a basic level. The report’s authors, among them the late secretary of education William J. Bennett, attributed this to “power over our education system [having] been increasingly concentrated in the hands of a few who don’t really want things to change, not substantially, not in ways that would really matter.”

Ironically, early proponents of ed tech arose partly in response to this bureaucratization of education. Where Arendt called for a restoration of authority, however, 21st-century technological “reformers” called for a project of liberation: the liberation of the student from the physical classroom, and the liberation of information from the authority of educators. Hence the rise of platforms like the Khan Academy, an astonishing start-up that has become perhaps the most influential ed-tech program.

In 2004, the MIT graduate and financial analyst Sal Khan began tutoring his cousin Nadia in math over the Internet, using Yahoo!’s Doodle Image service. In response to other friends and relatives seeking his help, Khan posted a series of math tutorials to YouTube in 2006. By 2008, the popularity of his videos prompted Khan to quit his job and found Khan Academy, an online ed-tech platform billed as a “personalized learning resource for all ages.”

Khan presented his new venture as the beginning of an education “information revolution.” Personalized lessons and feedback, among other features, would grant students newfound “agency over their own learning.” By providing tailored learning dashboards, practice exercises, and instructional videos, Khan Academy promises that pupils can “study at their own pace, both in and out of the classroom.” Information about the world—the information that educators are trained to master and impart—could be communicated to anyone, anywhere, by digital platforms at scale. Students would become sovereign learners, forever one click away from the very best of human knowledge.

Right from the start, Khan’s attempt to “liberate” students from the burdens of the classroom was motivated by a deep skepticism of the ability of teachers to offer meaningfully personalized learning. In traditional school settings, with twenty or thirty students but only one teacher per class, information is necessarily imparted to a collective—a mass of students who all receive the same lesson at once. For Khan and other ed-tech pioneers, this mode of learning would never succeed in catering to the needs and preferences of individual students, whether struggling or thriving. Of course, the traditional classroom format was never designed to be highly tailored to the individual; the classroom experience was social, allowing students to observe different styles of learning and gain patience or insight or kindness or boldness. To Khan, this constituted an inherent shortcoming.

Despite their disdain for the traditional classroom, however, Khan and other ed-tech entrepreneurs echoed Arendt in arguing that the bureaucratization of education had made things worse. Old-fashioned teachers might’ve been bad at personalized learning, but modern bureaucrats were downright lousy. The sovereign classroom for which Arendt expressed nostalgia was sub-optimal, but the bureaucratic, homogenized classroom of the late 20th century was fundamentally insulting to the dignity of students.

The unspoken premise of Khan’s case for “personalized learning” was that ed-tech companies would become the most effective guardians of true education. Teacher-led learning was somewhat attentive to individual students, but it could not scale. Administrator-led learning could scale, but not without flattening the experience of students. Ed-tech platforms would offer the best of both worlds: tailor-made lessons that catered to the needs and preferences of individual students while also delivering information to the masses, on an equal playing field. By combining individual learning with scale, ed-tech platforms could liberate students from the authority of both teachers and administrators. This liberation, though, came with a new form of dependency: students now must rely on technology companies for their studies, without meaningful intermediaries.

In an additional irony, Khan’s dream would never have become a reality without the active participation of the very teachers and administrators whom he sought to displace. The groundwork for his experiment had been laid some years prior to the launch of Khan Academy, when politicians and nonprofit leaders began employing the language of “digital citizenship.” While campaigning for the 2000 presidential election, George W. Bush and Al Gore both proposed large investments to prepare students for a “new age.” That same year, then-Governor Angus King of Maine announced a plan to provide laptops or comparable devices to all middle schoolers in the state. King promised to “demonstrate the power of one-to-one computer access . . . to transform education,” insisting that “the economic future will belong to the technologically adept.”

Across the country, these initiatives multiplied. Partnerships between state governments, ed-tech companies, and philanthropies invested hundreds of millions of dollars into personal devices for children. Apple’s ConnectED program, for example, aimed “to put an iPad into each student’s hands.” Google promoted its Chromebook devices and secured contracts with school districts. By the end of the 2010s, almost every school in America had provided devices for students—and the few that resisted were forced to yield when the coronavirus pandemic hit the nation in 2020. COVID accelerated the ed-tech revolution, and established the presence of screens in almost every American classroom. During the 2022–2023 school year, 94 percent of U.S. public schools claimed to provide students with personal digital devices such as laptops or tablets.

Armed with this technology, schools could partner with ed-tech companies to devise “digital learning” curricula and thus meet the demands of the Internet age. At first, these efforts were modest—students would spend a few hours watching lectures on Khan Academy or an equivalent platform. But ed-tech companies had every incentive to go much further, and Silicon Valley foundations were more than happy to help. In the past fifteen years, the Chan Zuckerberg Initiative—the philanthropic guardian of Mark Zuckerberg’s Facebook fortune—has spent upwards of $100 million on “personalized-learning” initiatives. Other Silicon Valley foundations, such as the Michael & Susan Dell Foundation and the Bill & Melinda Gates Foundation, have dedicated billions to similar projects. These investments accelerated the development of algorithmic learning software to teach reading, writing, history, mathematics, and foreign languages. Once confined to computer class—and used in limited ways, as a resource for acquiring digital skills like coding—tablets and laptops began to invade every classroom, absorb every subject, mediate every activity, and interject themselves between students and teaches, teachers and parents, and parents and administrators.

The relationship among ed-tech companies, Silicon Valley foundations, school bureaucrats, and state governments isn’t all that difficult to decode. The Internet economy grew by double-digit percentages every year after 2009; technology companies and friendly foundations were brimming with both cash and convenient optimism. As for education bureaucrats, technology promised not only a novel mechanism for top-down improvements in student outcomes, but also the ability to monitor and measure those potential improvements. Demonstrable results of bureaucratic success could, in turn, generate political plaudits and more money from foundations, state governments, and the Department of Education. Over time, school administrators became ed tech’s closest ally.

Sal Khan and other ed-tech pioneers envisioned their platforms as alternatives to the classroom, against the bureaucratization of education. But administrators understood that bringing ed tech into the classroom would produce the opposite effect, namely, accelerating the drive toward measurable, manageable, and transparent results that could be easily compared and efficiently analyzed.

In 1958, Arendt lamented the ever-increasing influence of experts over curricula, testing, and pedagogy. But this had happened in the open, often in explicit clashes between teachers, on the one hand, and progressive innovators and the bureaucrats who championed their policies on the other. Now, with the help of ed tech, administrators can achieve unparalleled levels of control without direct coercion. They no longer need to monitor teaching practices directly, since the platform dictates the terms of teaching in the classroom. They no longer need to monitor student evaluations, since the platform examines students on their behalf. In short, behavior no longer needs to be standardized or analyzed by individual administrators: bureaucratic platforms do that far more efficiently. Administrators get to entrench their own influence while speaking the language of “progress,” “innovation,” “transparency,” and “personalized learning.” Students have indeed been “liberated” from their teachers, but emphatically not from administrators, who wield more power over education than ever. Ed-tech, in other words, has become a powerful weapon in the hands of the very people it once sought to undermine.

Of course, the idea of “liberating” students from all authority was always folly. Khan wanted children to “exercise agency over their own learning,” as if ten-year-olds could design a curriculum that they would pursue with no help but that of Khan’s own program. This dream of self-sufficient children in charge of their own education seems absurd, but it is an application and extension of the view that children are “mini-adults” oppressed by the rules that govern them. But in fact, even with the most sophisticated of iPads in hand, children are not fully formed. They remain children—vulnerable, messy, susceptible to manipulation, and therefore in need of guidance. For most of them, learning what the curriculum requires isn’t a high priority. It is thus no surprise that weakening the authority of teachers by no means created a world of boundless self-expression. If anything, it achieved the exact opposite: schools that surveil their students and control behavior more closely than Arendt could possibly have imagined in the 1950s.

Teachers, meanwhile, are victims of these trends as much as children. No longer in charge of their own classrooms, they have become mere supervisors tasked with imposing discipline on the relationship between students and their devices. In other words, teachers are responsible for the implementation of the very technologies that erode their authority. “I feel like a cog in a machine,” one distressed educator told us. “I’m not in control,” another added. As Arendt wrote, the bureaucratization of the classroom has come “at the cost of the teacher’s authority,” creating “a situation in which the adult stands helpless before the individual child and out of contact with him.”

These aren’t isolated sentiments. In a recent UNESCO report surveying the consequences of the widespread adoption of education technology, teachers “expressed alarm at realizing they were, in some contexts and instances, tasked only with prompting students to progress down paths prescribed by automated systems.” Teachers feel powerless, and justifiably so—their role has been reduced to that of a technology facilitator. When asked about this sense of impotence, one teacher admitted: “Every year I teach, I feel like my job becomes more about monitoring and maintaining students’ attention than actually teaching anything to them.” Teacher doesn’t really capture the job description anymore.

The new role of teachers, that of facilitating technology use, has proved as challenging as it is demeaning. Most children (and many adults) view their devices first and foremost as portals to pleasure and entertainment. As the neuroscientist Jared Cooney Horvath writes, “digital technologies so often aren’t used for learning that giving students a laptop, tablet, or other multi-function device places a large (and unnecessary) obstacle between the student and the desired outcome; . . . students must expend a great deal of cognitive effort battling impulses that they’ve spent years honing—a battle they lose more often than not.” Teachers are therefore tasked with an uphill battle, watching over the shoulders of students as they inescapably gravitate back towards the countless distractions that their iPads or laptops offer.

Worse still, ed-tech companies are currently at work on automating this “supervisory” role as well. In school districts around the country, these corporations have already introduced comprehensive systems of control and surveillance. First, school districts have installed programs like GoGuardian—used by the Los Angeles Unified School District in the rollout of their 1:1 iPad program—on students’ devices. These programs have proved capable of monitoring, filtering, and supervising the digital activity of students more efficiently than the typical teacher. All these data about students are then combined and analyzed by other software programs to “create a profile for each student . . . relating to students’ health, wellness, and safety,” in the words of the ed-tech company Impero Software. For parents, the fact that education administrators are granted access to data about the capacities, behavior, preferences, and learning style of their children is unsettling enough. But this abundance of data also has worrying security implications: a whole generation is one cyberattack away from strangers knowing how they learned, behaved, and thought at a young age. As the autonomy of teachers continues to decline, these surveillance programs become ever more necessary. The more digital devices undermine the authority of educators, the more these programs seem like the only way to fill the disciplinary void.

In fact, to the extent that teachers still exercise authority in the classroom, education technology is increasingly deployed to make sure they do so in the right way. Hidden behind a candy-colored interface, the ed-tech platform ClassDojo asks teachers to monitor and record the behavior and skills of students by assigning them “Dojo points.” Students receive both positive and negative feedback, expressed through “pleasant” green dings and “harsh” red buzzing sounds. The platform tracks performance of both individual students and whole classrooms, before packaging it into reports for school administrators to review. Compared to more intrusive alternatives, ClassDojo promises a “softer option” for teachers who want a digital substitute for the old-fashioned report card. But even these “softer options” contribute to the creation of a permanent digital record with immense privacy and security risks.

This expansive rollout of surveillance and “management” programs notwithstanding, student discipline does not appear to be improving at all. In a recent survey, nearly half of teachers and administrators report that student misbehavior has increased over the last five years. Since the pandemic, the number of students chronically absent from school has nearly doubled. Many thousands of devices have themselves been reported stolen or missing; during the 2021–22 school year, over 77,000 school-issued laptops, iPads, or hotspots were lost or stolen from Chicago’s public schools alone. This amounted to over $23 million of lost technology.

In other words, even if we grant that student privacy can be sacrificed on the altar of “efficiency” or “transparency,” ed-tech platforms still fail on their own terms. It’s quite likely that they are exacerbating the underlying problems they are meant to solve. Far from enabling subtle but effective forms of discipline, our technology-powered education system has become both more intrusive and weaker.

The disempowerment of teachers—if not the downright destruction of teaching as a vocation—might be worth the price if it offered a better education for children. But ed tech has failed to improve educational outcomes across the board.

III. The Depersonalization of Personalized Education

Marshall McLuhan famously writes that “the medium is the message,” a catchphrase that captures the intuition that every mode of expression contains its own logic, rhetorical meaning, and atmosphere. Each medium favors certain ways of thinking at the expense of others. Put differently, how we say things shapes what we say. The Holy Roman emperor Charles V once confessed that he preferred to “pray in Spanish, socialize in Italian, woo in French, and admonish in German,” a 16th-century version of McLuhan’s aphorism. It’s hard to isolate the specific features that make French a more romantic language or German a sterner one. Yet these differences are there, imperceptibly constraining the terms of our encounters with each other.

Technology is no different. As with languages, each medium of communication remakes the world in its image. Before the printing press, education largely consisted of oral transmission. In the Middle Ages, Thomas Aquinas’s fellow students at the university of Paris had to memorize hundreds of texts by heart. These practices, constrained by the technologies of the day, favored a certain way of thinking about education. To educate was to receive, to absorb, and to recite. Oratory mattered a great deal, as did deference to masters on whom students depended for access to knowledge. The printing press upended this vision of education by democratizing access to the written word, and therefore to human knowledge. Oratory still mattered, but not as much. Students still memorized texts, but not as many or as systematically. In short, the terms of education and the role of teachers were both altered by technological development. The mass-printed book changed the face of education.

In Jewish tradition, too, the spoken aspect of learning is central. For much of Jewish history, the “Oral Law,” the vast collection of wisdom now contained in the Talmud and related texts, was not written down. This body of tradition was transmitted by word of mouth, from master to pupil. Manually recording the Oral Law was understood to be prohibited; it was also understood that some inherent wisdom would be lost should this material be transcribed. When the Talmud was finally committed to the page, it was not considered a joyous event, but a hard necessity.

Screens, too, represent a rupture in educational tradition. The digital medium reconfigures our activities in its own language—the language of data. Each activity becomes a series of discrete steps to be measured and “optimized.” Education is reduced to a set of “learning objectives” that ed-tech platforms pursue relentlessly. Learning a foreign language entails demonstrating the acquisition of x new words per week. Learning mathematics means solving a specific problem in under y seconds. Learning to write becomes about learning z grammatical rules per month. Thus every discipline must be understood through the need for metrics and quantifiable goals. Many digital platforms translate these goals into exercises that are algorithmically optimized for the individual student, whose performance is recorded in real time. Each student gets a personalized experience, carefully designed and continuously updated to maximize engagement. In the ideal digital classroom, the platform continuously monitors the student’s attention, response rate, and learning style. Each detail becomes an occasion for the platform to collect more data and devise an ever-refined product.

There’s a way to present this new reality as a utopia. After all, it’s but a logical extension of the bureaucratization of education. Demonstrable results, captured in quantifiable metrics. Full transparency, with both schools and parents checking on progress in real time. Personalized learning that nonetheless serves common standards of achievement. Self-improving experiences that capture the attention of students by “gamifying” every moment. By its own standards, ed tech has successfully optimized mass education. Students today are one iPad away from a learning experience that is more attentive to the needs of the individual student than the aristocratic tutors of pre-modern Europe. In what sense, then, does ed tech fail to educate children?

Part of it comes from the conception of education on which ed tech depends. Seen through the eyes of digital platforms, education becomes a series of discrete steps to be measured and optimized, a checklist of skills that algorithms translate into learning experiences. This way of thinking neglects the embeddedness of education. The act of learning is not merely about the acquisition of quantifiable skills, but also about the development of the self and the cultivation of relationships. A teacher is not an impersonal vessel of knowledge, but a person whom students encounter, with a story and a voice of her own. In the classroom, children cultivate qualities that cannot be measured by an algorithm. Would anyone entrust digital platforms with the development of artistic taste in children? Or of ethical responsibility? Or of curiosity, creativity, or emotional intelligence? No doubt there are some enthusiastic engineers working to do just that. But try as they might, these are qualities that ed-tech platforms cannot quantify or “optimize”—virtues embedded in precisely the kinds of activities that ed-tech platforms undermine.

Consider handwriting as an example. If handwriting were merely a way of communicating information in written form, students would not lose much by learning to type on a computer or a tablet instead. But it isn’t. The neuroscientist Karin James has shown that “handwriting is important for the early recruitment in letter processing of brain regions known to underlie successful reading.” Comparing five-year-olds learning how to write by hand to five-year-olds learning to type instead, she found that the first group awakened entire parts of the brain that remained dormant among the second group. Her MRI scans provide scientific evidence of what most of us know intuitively: handwriting cultivates a certain relationship to the written word that typing cannot replicate. The slow, patient, careful carving of letters develops our minds in a way that the frenetic hammering of keys does not.

This study is far from the only one to show the benefits of traditional handwriting. Evaluating the performance of 700 elementary-school students who were struggling with English, the psychologist Virginia Berninger found that those who spent more time handwriting systematically outperformed the others—not just at reading and writing, but also at memorizing and expressing creative ideas. Perhaps most importantly, those who focused on handwriting reported enjoying learning more. In a subsequent study, Berninger even found that learning how to write in cursive improved the ability to visualize concepts and ideas. Like James, Berninger demonstrated that writing by hand awakens a “region of the brain, the fusiform gyrus, where visual stimuli actually become letters and written words.” For Berninger, handwriting is a gateway to our imagination—a portal between the mind and the hand for which typing, a more disembodied experience, offers no substitute.

Scientists have observed the same patterns when it comes to drawing by hand, learning mathematics the old-fashioned way, or reading physical books instead of digital ones. In each of these cases, the pre-digital activity has manifold benefits that its digital substitute lacks. Drawing by hand, for instance, is among the few activities to create pathways between the two hemispheres of the brain—the right side for creativity, the left side for logical thinking. It relieves stress and improves our spatial awareness better than scribbling on an iPad or using an image-generator can. Similarly, dozens of studies have demonstrated that traditional mathematics—mental calculations, with as little help from calculators or computers as possible—is not only better at developing quantitative reasoning, but also at cultivating a wide range of cognitive abilities that aren’t just relevant to mathematics. Shortcuts, as J.R.R. Tolkien put it, make long delays.

In fact, the core shortcomings of screens in classrooms aren’t subject-specific at all. The psychologists Pam Mueller and Daniel Oppenheimer, for instance, have shown that taking notes on a computer is systematically worse than doing so by hand, both for the person taking the notes and for other students nearby. No matter the subject, taking notes on a computer means retaining less information and engaging with the material less critically.

Part of this comes from the simple fact that laptops are full of distractions—students are always a few clicks away from the news of the day, texts from their friends, or social media. But Mueller and Oppenheimer go even further, showing that even “when laptops are used solely to take notes, they may still be impairing learning because their use results in shallower processing.” In other words, even if the students don’t get carried away by the distractions at their disposal on the screen, their learning outcomes remain worse across the board. And that’s because the medium itself stands in the way of genuine education.

Psychologists and neuroscientists can give us a sense of what’s missing, but they cannot capture the deeper damage that ed tech inflicts on our children when it robs them of embedded experiences. Even the most sophisticated of MRI scans cannot convey the connection between mind and body that handwriting forges. “It’s not just that [our children] don’t write as well,” one worried parent told us. “It’s that they don’t seem to think through writing in the same way.”

Writing with a pen binds our thoughts to our hands, submitting our internal monologue to the rhythm and constraints of word-crafting. Handwriting, drawing, and mental mathematics are indeed “inefficient,” in the way that all embodied experiences are inefficient: these activities force our mind to confront the limits of the flesh, a friction out of which a stronger sense of self emerges. That sense of self, more than any skill or virtue, is what screens undermine. Handwritten notes are our own in a way that typed notes are not. They express the idiosyncrasies of the mind through the body, fusing the two without intermediaries. To feel at home in one’s body is to cultivate and cherish that kind of connection. To forego that kind of connection is to condemn children to self-estrangement, to teach them that their bodies are but uncomfortable burdens to be discarded when something better comes along.

Ed-tech companies are fully aware that their platforms harm the ability of children to concentrate, memorize, and write creatively. That’s why they invest immense amounts of money into studies that purport to show the benefits of their own products. That’s also why they dedicate enormous resources toward making their platforms as “engaging”—or, less euphemistically, addictive—as possible.

Engineers speak of “gamification” to describe the assumption that learning must be entertaining—that education should mirror the curated experiences that students encounter on their phones. As with videogames or social media, ed-tech engineers design every detail of their interface to retain the user’s attention. Every math problem becomes a game, every language lesson an adventure. The platform’s logo, color scheme, layout, typography, and interactive features all exploit human psychology to glue students to the screen. Just as teenagers spend hours scrolling through Facebook without realizing it, so ed-tech platforms do everything in their power to make the experience of education as seamless, and absorbing, as possible. The platform, which collects copious quantities of data in real time, makes itself more and more “engaging” as time goes on. “Personalized learning” does not merely mean exercises adapted to the needs and abilities of the child, but also an interface tailor-made to exploit every child’s mental reflexes. While these technologies can constantly adjust to respond to the feedback they collect from the student, they are in fact depersonalizing—reducing the child to a set of data.

For those who see education as a checklist of skills, there’s nothing wrong with gamification. Teachers have always tried to keep their students focused, and ed-tech platforms allow them to do that more efficiently than ever. Yet this view fundamentally misunderstands both the nature of genuine learning and the difference between engagement and addiction. Real education requires wrestling with ideas and developing the capacity for sustained attention—exactly the skills that digital learning undermines in the first place. Instead of cultivating the right approach to learning, ed-tech platforms decimate the ability of children to concentrate and then create the illusion of “engagement” by making themselves as addictive as possible. Can we truly call students “engaged” if the only thing that keeps them going is the algorithm exploiting their minds behind the scenes? Should children be “engaged” in class in the same way that they are “engaged” when they mindlessly, near-helplessly watch videos on TikTok? Gamification does not get students interested in the topics they study. At best, it produces entertainment masquerading as passion. At worst, it hides the destruction of our children’s ability to concentrate, for which screens are themselves responsible, behind a flashy interface without which students would be incapable of focusing for more than half an hour.

Ultimately, the reduction of education to a combination of entertainment and metrics mirrors similar transformations across society. Just as social media optimize for engagement rather than meaningful connection, ed-tech platforms optimize for completion rates and test scores rather than genuine understanding. These platforms decimate the cognitive abilities of children, and then keep children artificially “engaged” by means of gamification. In other words, just like Facebook or X, ed-tech platforms benefit from the harm that they inflict on younger generations. The more students learn in front of a screen, the more students must learn in front of a screen, for they have lost the ability to do otherwise.

The result is an entire generation incapable of writing without a keyboard, adding without a calculator, drawing without an image-generator, paying attention without a curated interface, and—soon enough—thinking without ChatGPT. To say that these children depend on technology is to indulge in comforting euphemisms. In the 1990s, the pioneers of ed tech envisioned a generation of “digital natives” who would thrive in “multimedia environments” and express themselves in code. That myth has been thoroughly debunked—today’s students are no more capable of productive multitasking than their parents were, but they’re worse at just about everything else.

IV. How to Save Education

The rise of artificial intelligence is rapidly raising the stakes of the ed-tech debate. A few weeks ago, the California State University system announced a partnership with OpenAI to launch the largest deployment of artificial intelligence in higher education to date—23 campuses, serving nearly 500,000 students, will integrate ChatGPT into their curricula and operations. Public and private high schools will likely follow, and this kind of initiative might even reach primary schools down the line.

In the not-so-distant future, children will be taught to write with AI, count with AI, and think with AI. Behind the scenes, AI will also improve the ability of ed-tech companies to “personalize” learning and keep students “engaged” by processing data en masse. If current trends continue, ed tech will penetrate more and more facets of education, collect more and more information on students, and become more and more addictive. As the high-school teacher Daniel Herman teacher recently argued in the Atlantic, it’s no exaggeration to say that AI chatbots might herald the “end of high-school English.” And this, too, will be defended as a form of “liberation” for children.

Fortunately, some schools are starting to resist. New York’s education department, which has agreed to large-scale partnerships with ed-tech companies in the past, recently decided to block ChatGPT on school devices and networks. At the moment, this policy is an outlier. But it shows that there is some appetite for questioning the hegemony of ed tech over our education system—an appetite that could grow as the damage becomes more evident.

Unfortunately, public schools might not prove capable of resisting the pull of ed tech, if only because of the perverse incentives tying school administrators to Silicon Valley foundations and technology companies. Some school districts have signed decade-long contracts enshrining the dominance of ed tech for years to come. Others will struggle to justify spending millions of dollars on tablets and software that they’re no longer using. Even with the most effective of awareness campaigns, this web of incentives and obligations might tie the hands of public-sector reformers in the short to medium term.

Private schools, on the other hand, are in a much better position to act—but they must act together, at scale. The efforts of the Jewish Parents Forum to ban phones in Jewish day schools offer a blueprint for collective action. Thanks to Bryk, Haidt, and others, Jewish schools across the country have already begun to redefine their relationship with technology. Many of these educators and parents are asking the right questions. Which skills truly require digital tools to develop? When does technology enhance learning, and when does it detract from it? How can we prepare students for a digital world without surrendering the entire educational experience to screens? The answers won’t be simple, but they start with reclaiming agency. Schools need to come together to rethink their relationship to ed-tech companies, insisting on tools that serve their educational philosophy rather than replacing it.

Jewish schools can also draw on distinctly Jewish traditions of learning in answering these questions. First is the idea of the teacher as a repository of tradition, and textual authority, rather than as a handmaiden of the algorithm. Traditional Jewish study is also intensely social, not an atomized activity but one that involves encountering other points of view and sharpening one’s own. Walk into any yeshiva and you will see students sitting in pairs over written texts, noisily debating and interpreting—and nary a screen in sight. The talmudic tractate Ethics of the Fathers enjoins scholars to “appoint a teacher and acquire a friend,” which is usually taken to mean that both an instructor and a study partner are necessary to learn. The Talmud is replete with stories of pairs of such study partners, or havrutot, who owed virtually all the wisdom and knowledge they attained to one another, and who discovered new insights in the common pursuit of wisdom. Moreover, this relationship, like that between master and pupil, is at its best meant to be moral and social, not merely academic. Real learning in this tradition requires long hours of focus, combative friendships, and ethical guidance.

In recent decades, we’ve conducted a vast experiment on a generation of children, replacing time-tested practices with digital alternatives without understanding their full implications. The solution isn’t to abandon technology entirely—that ship has sailed, and even if it hadn’t, some of these innovations could be beneficial if used wisely. Instead, schools need to draw clearer boundaries between digital literacy and the core curriculum, ensuring that screens enhance rather than undermine the activity of learning. Otherwise, we risk raising a generation that knows how to interact with screens but has never learned to think deeply, relate personally, or engage meaningfully with the world around them.

Teachers have a clear interest in regaining control over their classrooms. Parents don’t want schools to worsen their kids’ addiction to screens instead of limiting it. Even administrators would benefit from fewer costly subscriptions to services that actively harm students. Unlike in public schools, the incentives of all parties involved are aligned. But the past few years have shown that mere awareness and agreement aren’t enough to act. This is where institutions such as the JPF come in, as mediators who gather parents, teachers, and administrators within a community to build consensus around new policies. From one school to the next, from one community to the next, from one state to the next, the fight over the future of ed tech will take time, effort, and resources. The Jewish community is uniquely positioned to get this movement started, if only because the weekly practice of Shabbat gives us a sense of what it means to separate the needs of a technological world from spaces in which that world has no place.

As Abraham Joshua Heschel puts it in his book on the Sabbath, “the solution of mankind’s most vexing problem will not be found in renouncing technical civilization, but in attaining some degree of independence of it.” Should Jewish schools succeed at doing that, the rest of the country might follow.