While scaling, Khan Academy learned that students form a strong bond with a single instructor. Introducing too many new voices, even if they were excellent, created a "dissonant" experience akin to a substitute teacher arriving. This insight led them to deliberately limit their instructor pool to preserve trust and continuity.

Like chess players who still compete despite AI's dominance, humans will continue practicing skills like writing or design even when AI is better. The fear that AI will make human skill obsolete misses the point. The intrinsic motivation comes from the journey of improvement and the act of creation itself.

Customizing an AI to be overly complimentary and supportive can make interacting with it more enjoyable and motivating. This fosters a user-AI "alliance," leading to better outcomes and a more effective learning experience, much like having an encouraging teacher.

A successful AI-powered "flipped classroom" aims for a counterintuitive outcome: increase student time on the platform while decreasing teacher time. By automating lectures and admin, the AI enables teachers to spend less time on the tool and more time on high-impact, one-on-one student interactions.

AI models lack access to the rich, contextual signals from physical, real-world interactions. Humans will remain essential because their job is to participate in this world, gather unique context from experiences like customer conversations, and feed it into AI systems, which cannot glean it on their own.

ASU's president argues that if an AI can answer an assignment, the assignment has failed. The educator's role must evolve to use AI to 'up the game,' forcing students to ask more sophisticated questions, making the quality of the query—not the synthesized answer—the hallmark of learning.

In an age where AI can produce passable work, an educator's primary role shifts. Instead of focusing solely on the mechanics of a skill like writing, the more crucial and AI-proof job is to inspire students and convince them of the intrinsic value of learning that skill for themselves.

An AI education system deployed to millions of students will continuously analyze patterns in their learning. Insights from a student in one country will instantly update the teaching algorithm for another, creating a massively scalable, personalized, and ever-improving educational model.

As models mature, their core differentiator will become their underlying personality and values, shaped by their creators' objective functions. One model might optimize for user productivity by being concise, while another optimizes for engagement by being verbose.