Responsible AI adoption in education requires two things. Both are necessary. Neither is sufficient alone.

The first is pedagogy. Does the tool strengthen the student-teacher-content relationship, or weaken it? Does it support the cognitive work that produces learning, or displace it? Brookings, UNESCO, and UNICEF have devoted serious resources to these questions. AI enriches learning only under specific conditions, and those conditions require deliberate design.

The second is governance. Who controls the environment in which pedagogy occurs? What has the institution agreed to before the tool ever reached a classroom? Does it retain the capacity to change direction, or has it ceded that capacity to a vendor without realizing it?

These are equally the right questions. The field has not been asking them at anything close to the same scale.

No editorially independent publication in education connects the research community to itself and to the practitioners whose work it is supposed to inform — across formats, across disagreements, and including critical coverage of its own structural conditions.

Neuroscience built that infrastructure when it created The Transmitter and its recurring series "How to Teach This Paper" — a model for walking practitioners through not just what a study found, but how it was designed, what the methods assume, and what it would take to apply it responsibly. Education has not.

The problem is not the absence of data. It is the inability of systems to turn evidence into instructional improvement. Two reports. Two contexts. One structural failure.

This week in UNBOUND, a study of data storytelling in U.S. data science classrooms and a policy brief on foundational learning data in Kenya reveal the same challenge: education systems often generate evidence but fail to translate it into changes in teaching.