Students are the chef of AI and AI is just a capable line cook

Right now, I teach highly motivated students, but many of them have their sights firmly set on STEM careers. That's wonderful, but I often see a tendency to treat subjects like history or geography as just boxes to check. Something to get through, rather than a vital way to develop critical thinking. Naturally, when GenAI is available, the temptation to use it just to 'get the job done' is huge. They risk missing the conceptual skills and intellectual depth that historical thinking concepts or geographical inquiry adds, which are skills that truly enrich any future path, STEM included.

GenAI is a powerful game-changer and it's transforming society fast. The problem is that our education system simply hasn't caught up to where our students already are. Board policies, at least here in Ontario, often aren't much help. They tend to focus heavily on what teachers shouldn't do, giving us vague, nervous instructions instead of clear guidance on how we can actually integrate and leverage this technology. That leaves a lot of us feeling exposed and, crucially, leaves students unsupervised in how they choose to use the tools.

So, I’ve shifted my focus entirely. I've moved past trying to police GenAI and instead focus on teaching students to use it as a powerful ally. I emphasize that students are the "chef," and AI is just a highly capable "line cook" in their learning kitchen.

This means putting the responsibility back on them:

• The Chef Sets the Menu: They have to clearly understand the goal—what argument, concept, or historical perspective they want to achieve—before they ask AI for anything. This step forces upfront intellectual engagement.
• The Chef Oversees the Cooking: They need to be masters of prompt engineering, refining and directing the AI to produce something that reflects their unique voice, specific assignment criteria and final intent. This process sharpens their critical thinking and subject-specific communication skills.
• The Chef Taste-Tests: The final step is rigorous editing and verification. They must check the output, making sure it accurately captures their ideas and demonstrates their authentic understanding. Often times the output needs a bit of extra seasoning.

If we successfully teach students how GenAI can make them better, smarter and more efficient thinkers, they hopefully won’t need to rely on it as a mental crutch. Instead, they will harness its power without sacrificing their ability to think for themselves.

The biggest practical issue we all face is simply the 24/7 nature of access. We can’t watch them all the time and stopping all use would require draconian penalties that would throw away the instructional opportunity we have. Our only realistic strategy is to keep guiding them, hoping that the positive skills we teach will outweigh the temptation to misuse the tool.