Two sectors
K-12 vs Higher Ed
K-12 evidence comes from Dunnigan et al. (2023), with teacher-education implications from Henriksen et al. (2026). Higher-education evidence comes from Chan (2023) and Hughes et al. (2025).
How is the problem defined?
K-12
No map yet, so policy can wait. “Generative AI doesn't come with a user manual.”Higher Ed
Integrity and readiness need an institutional plan (Chan); nine interlinked challenges (Hughes et al.).Who should act?
K-12
Leaders look up and out: the state, universities, and industry.Higher Ed
Teachers, senior management, and IT staff each own a dimension; HE decision-makers and regulators.Policy status
K-12
With one exception, no AI-specific policy. Existing acceptable-use policies “cover it.”Higher Ed
Students (4.50) and teachers (4.54) strongly want a plan for managing risks now.Training
K-12
No urgency. Post-COVID change fatigue; curriculum and testing come first.Higher Ed
AI literacy is Key Area 6; knowledge disparity is a top driver of all other challenges.Equity
K-12
Raised by one leader in three hours. Paid tiers open “a new chasm in the digital divide.”Higher Ed
Universities should procure tools so access doesn't depend on income; divides within cohorts and across countries.What is at stake
K-12
Teachers' fears: “Is this gonna replace me? How am I gonna know if my students are doing authentic work?”Higher Ed
Academic standards, the traditional academic role, and the validity of the degree itself.Bridging both
Teacher education sits in between
Principled engagement
Move from tool training to principled engagement. Practices built on compliance or rote reproduction are the ones AI copies.
Design for process
Build in deliberation, revision, and metacognition. Struggle is a condition of learning.
Foundations first
Conceptual understanding, disciplinary grounding, and ethical judgment make meaningful AI use possible.
Shape futures
“The future is what we make of it.” (Dede) Educators become educator-designers.