Recent usage data released by OpenAI shows a genuinely staggering figure: US classwork-related prompts now exceed 460 million per week, alongside roughly 70 million weekly knowledge-check prompts specifically. This is a single provider's own data, representing just one part of the broader AI landscape students interact with, which means the true total volume of student AI use for schoolwork across all providers combined is almost certainly considerably higher still. Whatever the exact total, this scale confirms something worth taking as a settled premise for classroom planning: AI use for schoolwork isn't a marginal, occasional phenomenon among a small group of students, it's a genuinely massive, routine part of how a very large share of students now approach their academic work.

This volume has real, practical implications for how classroom policy and assignment design should be approached. A policy built on the assumption that most students aren't using AI for their work, or that AI use is an edge case worth addressing only occasionally, is working from an outdated premise given data at this scale. The more realistic starting assumption is that a meaningful share of students are already using AI tools for at least some part of most assignments, which shifts the practical question from whether to address AI use to how to address it thoughtfully and specifically.

This scale also reinforces why assignment design that assumes genuine, class-connected engagement, rather than assignment design that assumes AI use is rare and easily deterred, represents the more realistic, sustainable approach going forward.

What this volume means for realistic classroom planning

Given data at this scale, the most realistic and productive classroom approach treats AI use as a given, routine part of the current landscape, shifting focus toward clear disclosure expectations, assignment design that requires genuine personal and class-specific engagement, and transparent grading practices, rather than an approach built around the assumption that AI use is rare enough to be caught and deterred case by case through detection alone.

• Treat AI use for schoolwork as a routine, widespread current reality, not a rare edge case, when designing classroom policy
• Focus policy energy on clear disclosure expectations and assignment design, rather than detection-based deterrence alone, given this scale
• Design assignments assuming a meaningful share of students are already using AI tools for at least part of the work
• Build genuine, class-connected engagement requirements directly into assignments, since generic prompts are exactly what this volume of AI use handles most easily
• Revisit your own classroom AI policy periodically, given how quickly usage volume continues to grow

Why this reinforces transparency over detection as the primary strategy

At this scale of use, a detection-first strategy, trying to catch and address AI use case by case after the fact, becomes considerably less practical and less effective than a transparency-first strategy, building clear expectations and genuine engagement requirements directly into how assignments are designed and graded from the start. This reinforces a shift that's already been building across current academic integrity discussion this year: proactive design over reactive detection.

For grading specifically, this scale of AI use is also a useful reminder of why transparency about a teacher's own AI use in the grading process matters, given that students engaging with AI tools this routinely are likely to have real, informed expectations about how AI might reasonably factor into their own feedback and evaluation as well.

A scale worth taking as the new baseline

This kind of usage data offers a genuinely useful, concrete anchor for classroom planning this year: AI use for schoolwork is happening at a truly massive scale, and building policy, assignment design, and grading transparency around that reality, rather than around an outdated assumption of rare, occasional use, is the more realistic and productive path forward.