What We Get Wrong About AI and Education

What is Steve Hargadon's four-level framework for understanding AI in education?

Steve Hargadon distinguishes between schooling (institutional layer focused on conformity and credentialing), training (skill acquisition for specific purposes), education (classical sense of drawing out higher-level thinking), and self-directed learning (the ultimate goal of lifelong learning). He argues this framework clarifies why people have conflicted reactions to AI in education, as each level presents different opportunities and costs.

Why does Steve Hargadon say we think in binary terms about AI in education?

According to Hargadon, binary thinking about AI (either saving or destroying education) is a cognitive consequence of not having a proper framework for analysis. When we can only see the institutional layer of learning, we default to all-or-nothing thinking because the difficulty in understanding what's really happening leads us to gravitate toward simple, clear positions.

How does Steve Hargadon define the difference between schooling and actual learning?

Steve Hargadon argues that schooling is primarily an institutional sorting system that teaches conformity, rule-following, and performance on standardized measures, with its main output being credentials. He notes that students who excel at schooling often describe themselves as 'good at the game of school' rather than as good learners, highlighting the distinction between institutional success and genuine learning.

What does Steve Hargadon mean by 'definitional confusion' in institutions?

Hargadon describes how institutions inevitably collapse important distinctions within their domain because they need things to be uniform, legible, and measurable. He argues this isn't corruption but an inevitable consequence where activities that keep institutions alive and expanding generally don't serve the original mission, reshaping human needs into institutional responses.

How does Steve Hargadon compare education's definitional confusion to medicine and employment?

Steve Hargadon draws parallels showing how institutions conflate different concepts: healthcare systems treat medical procedures as synonymous with health, while employment systems treat job creation as the solution to economic insecurity. He argues education suffers the most harm from this dynamic because the stakes are personal and the gap between stated purposes and actual functions is particularly wide.

What are the specific costs of AI at each level in Hargadon's framework?

According to Steve Hargadon's framework, AI costs vary by level: in schooling, it breaks the credentialing game; in training, it risks producing shallow competence; in education, it offers a tool limited in human reasoning capabilities; and in self-directed learning, it creates risks of filter bubbles and flattened curiosity. He emphasizes the framework doesn't eliminate these negatives but brings them into sharp focus.

Why does Steve Hargadon say feeling both alarmed and excited about AI in education is reasonable?

Hargadon argues that conflicted feelings about AI in education reflect genuine responses to different frames of reference that people don't realize are different. He contends that both alarm and excitement are reasonable responses that can be navigated without choosing only one, as they correspond to different levels in his learning framework.

What does Steve Hargadon identify as the historic breakthrough potential of AI for self-directed learners?

Steve Hargadon sees AI as potentially offering self-directed learners unprecedented access to a responsive, patient, and knowledgeable interlocutor available at any hour. He argues this breakthrough collapses traditional barriers of geography, cost, and institutional gatekeeping that previously limited learning opportunities.