Can AI make every teacher better?

Most AI tools in education are trying to replace something.

Replace marking. Replace planning. Replace feedback. Replace the teacher.

I think that is the wrong frame entirely.

The most powerful use of AI in education is not replacement. It is augmentation.

Giving educators capabilities they have never had access to before, not because they lacked the expertise, but because they lacked the time, capacity, and infrastructure.

Here’s what I mean.

We already know that high-quality instructional coaching works. One of the largest meta-analyses ever conducted on teacher coaching, by Kraft, Blazar and Hogan (2018) examined 60 causal studies, and found substantial improvements in teaching practice and measurable gains in student achievement. The challenge was never whether coaching works. The challenge was scale.

An experienced instructional coach can observe a lesson and provide detailed feedback on questioning, explanations, classroom management, checks for understanding, and teacher-student interactions.

But even an excellent coach can only support a limited number of teachers each week.

A school with 60 teachers could wait months before every teacher receives meaningful observation and feedback.

This is where AI becomes interesting.

Not because it replaces the coach.

Because it makes the benefits of coaching available at a scale that was previously impossible.

AI can analyse patterns, surface evidence, identify strengths and areas for development, and provide immediate feedback after every lesson. The coach is then freed to focus on the work that matters most: interpretation, professional judgement, motivation, culture, and strategic improvement.

The coach becomes a strategist.

The AI becomes the infrastructure.

More broadly, this aligns with a growing body of research on human-AI complementarity, which suggests that the greatest gains often come not from replacing human expertise, but from combining human judgement with AI-supported analysis and insight.

Education has always struggled with a fundamental problem: the things that improve teaching most are often the hardest to scale.

AI gives us an opportunity to change that.

Not by replacing educators.

By extending what great educators can do.