Personalized learning has become a dominant force in educational reform, promising to revolutionize teaching and learning through technology-enabled “individualization”. Proponents claim it will improve student engagement and accelerate learning. However, a critical analysis of the outcomes reveals fundamental problems with both the concept and implementation of personalized learning. Despite its appeal and ambitious promises, personalized learning has failed to deliver on its potential because it narrows education to a set of measurable competencies, accelerates the commodification of education, and dehumanizes the learning process. As Pelletier (2023) argues, we need “nuanced, thoughtful and empirically-grounded responses” to questions about the role of technology in education (2023, p. 111). Instead, the discourse has remained uncritical, overlooking that personalized learning fails to produce better student outcomes.
One of the most fervent criticisms of personalized learning is how it reduces education to a series of discrete, measurable competencies. Brass and Lynch (2020) trace this reductionist approach back to behaviorist psychology and scientific management principles, showing how personalized learning descends directly from Skinner’s teaching machines, which were themselves derived from experiments training pigeons to guide missiles. This peculiar ancestry reveals how personalized learning perpetuates a mechanistic view of learning, where knowledge is broken down into “observable and measurable responses that were ‘simple and definite enough to permit handling of much routine teaching by mechanical means'” (Brass & Lynch, 2020, p. 7). The linear, incremental nature of edtech platforms ignores the complex, non-linear nature of genuine learning and understanding. As Pelletier (2023) observes, this represents a fundamental misunderstanding of education, noting that even “the phrase ‘learning at your own pace’ presupposes it has a beginning and an end point” (p. 113). This reductionist approach raises profound questions about the purpose of education itself: “What vision of a better society, a better economy, a better way of living together, does ‘personalised learning’ imply or promote?” (Pelletier, 2023, p. 113).
The commodification of education through personalized learning represents an even more insidious transformation of educational practice. Brass and Lynch (2020) reveal how modern personalized learning platforms emerge from a complex network of commercial interests, where “businesses, philanthropies, and governments have worked together to open the education market to for-profit and non-profit providers” (p. 13). This reflects what Arantes (2024) identifies as the broader “marketization of education systems” (p. 537), where learning becomes a product to be packaged and sold rather than a human developmental process. The commodification extends far beyond just the platforms themselves to encompass student data and learning processes. Ashman et al. (2014) detail how personalization technologies enable unprecedented collection and monetization of student data, creating new privacy concerns and ethical dilemmas about “the ownership of this personal data” (p. 6). The data collection serves commercial interests more than educational ones, as companies use it to “develop educational products that exploit patterns in users’ online activity” (Brass & Lynch, 2020, p. 15).
Perhaps the sharpest criticism of personalized learning lies in its potential to dehumanize education by diminishing the essential role of human relationships in learning. This dehumanization operates on multiple levels, from the reduction of teacher agency to the erosion of social learning opportunities. Dumont and Ready (2023) emphasize that “learning is more than a cognitive activity; rather, it encompasses multiple emotional, motivational and social processes” (p. 3). Yet personalized learning platforms often reduce teachers to mere facilitators or data monitors, undermining their professional judgment and ability to respond to students’ needs holistically. This transformation of the teacher’s role represents what Pelletier (2023) identifies as a fundamental misunderstanding of educational practice, arguing that truly personalized instruction already happens in traditional classrooms through the natural adaptability of skilled teachers. The shift toward algorithmic control in education raises serious concerns about autonomy and agency. Buchanan’s (2023) analysis represents a shift towards “control societies” where students become “dividuals”—collections of data points to be measured and managed (p. 4). Surveillance and control threatens to convert education from a human developmental process into a form of algorithmic management, where learning pathways are predetermined by software rather than shaped through genuine human interaction and professional judgment.
The lack of empirical evidence supporting personalized learning’s effectiveness is troubling, given its widespread adoption and promotion. Dumont and Ready’s (2023) comprehensive review reveals that “the emerging evidence for the benefits of personalized learning to reduce educational inequalities is much more mixed” than advocates suggest (p. 2). Their study shows that technology-enabled personalized learning may actually disadvantage struggling students, with research finding that “initially low-performing students and students with lower working memory capacity learned less than their more competent peers” (p. 2). Brass and Lynch (2020) offer historical context in their assessment of previous iterations of personalized learning, which were “deemed educational and commercial failures” (p. 17), yet similar approaches continue to be promoted without substantial evidence of improvement. Even when personalized learning shows technical promise, as Magomadov (2020) suggests in discussing AI applications, significant implementation challenges often undermine its effectiveness in practice. These findings suggest a troubling disconnect between the lofty promises of personalized learning advocates and the reality of its impact on student learning and educational equity.
The historical context provided by Brass and Lynch (2020) is particularly revealing, as they demonstrate how personalized learning represents not a revolutionary new approach but rather the latest iteration of repeatedly failed attempts to automate and commodify education. Their analysis traces how today’s personalized learning platforms share significant DNA with Pressey’s “Automatic Teacher” and Skinner’s teaching machines—technologies that were ultimately rejected for many of the same reasons critics raise today. Yet despite this historical pattern of failure, personalized learning continues to be promoted as a universal panacea for education’s ills, driven more by commercial interests and techno-solutionism than by evidence of effectiveness or sound educational theory. As Arantes (2024) argues, this reflects a broader trend where educational technology is “trending towards the augmentation or replacing human teachers with non-human technology” (p. 524) without adequate consideration of the implications and consequences.
While personalized learning continues to be promoted as an educational revolution, careful analysis reveals fundamental problems with both its premises and implementation. Its narrowing of educational goals, acceleration of educational commodification, dehumanizing effects, and lack of empirical support suggest that skepticism is not only warranted but necessary. As Markham (2021) argues, “the power of anticipatory logics and ‘trajectorism’ flowing through everyday discourse around technologies builds and reinforces a hegemonic ideology of external power and control” (Markham, 2021, p.384). Such a model risks reinforcing, rather than questioning, existing power dynamics within education—which is precisely why its promises should be met with careful and rigorous critique. The challenge moving forward is not to reject technology entirely but to develop approaches that maintain the human core of education while thoughtfully incorporating technological tools where appropriate. This requires maintaining what Ashman et al. (2014) call the “bidirectionality” of education: the genuine interaction between teachers and students that is essential to meaningful learning. Only through such an approach can education maintain its status as a public good, one that prioritizes the holistic development of individuals and supports the formation of an informed, empowered society.
References
Ashman, H., Brailsford, T., Cristea, A.I., Sheng, Q.Z., Stewart, C., Toms, E.G. and Wade, V., 2014. The ethical and social implications of personalization technologies for e-learning. Information and Management, 51(6), pp.1–24.
Arantes, J., 2024. Digital twins and the terminology of “personalization” or “personalized learning” in educational policy: A discussion paper. Policy Futures in Education, 22(4), pp.524–543.
Buchanan, R., 2019. ‘Through Growth to Achievement’: Education Policy, Personalized Learning, and the Control Society, Paper presented at the annual meeting of the American Educational Research Association, pp.1–6.
Dumont, H. and Ready, D.D., 2023. On the promise of personalized learning for educational equity. npj Science of Learning, 8(26), pp.1–6.
Magomadov, V.S., 2020. The application of artificial intelligence and Big Data analytics in personalized learning. Journal of Physics: Conference Series, 1691(1), pp.1–4.
Markham, A., 2021. The limits of the imaginary: Challenges to intervening in future speculations of memory, data, and algorithms. New Media & Society, 23(2), pp.382–405.
Pelletier, C., 2024. Against Personalized Learning. International Journal of Artificial Intelligence in Education, 34, pp.111–115.
Great work Melissa and not much to add here to the critique, which I felt coherent and convincing in what it was trying to do (perhaps best typified by your thesis statement in ‘Despite its appeal and ambitious promises, personalized learning has failed to deliver on its potential because it narrows education to a set of measurable competencies, accelerates the commodification of education, and dehumanizes the learning process.’
I thought the way you used the research to support your arguments rather than drown them out is convincing as well. Good use of the sources from the course and from your own digging around. All I think indicate a strong sense of critical awareness around how academic discourse is constructed and supported. Very well done indeed.
Although this wasn’t the question being asked, and so this is more rhetorical than anything else, but I wonder if there exists a conceptualisation of personalisation that might not narrow, accelerate, and dehumanise? Or perhaps put another way, is the fault of personalisation really around its inappropriate conceptualisation rather than the idea itself? Does a position of personlisation exist that may prove complementary to Biesta’s socialisation and subjectification? Is there a way to see the broader ‘teacher function’ as an assemblage of parts, one of which is personalisation? (Bayne, S. (2015). Teacherbot: interventions in automated teaching. Teaching in Higher Education, 20(4), 455-467.).
This is one of those examples (again outside the scope of the question being asked) where the critical vantage point one executes the critique is important. If we personalisation on its own, your critique is perfectly valid (again, I promise you did a good job here!). If we see personalisation as part of a larger teacher function, then any individual shortcoming of personalisation might be offset or accounted for somewhere else in that teacher function. In short, I am just kind of previewing the last block of the course where we move into a bit more creative a space and try to decide the optimal balance for preferable futures. Should be good fun!