Link: https://moztako.me/how-teachers-detect-ai-generated-student-work/

TIL that generative AI is changing how teachers assess student work.

Instead of looking only at the final assignment, some educators are paying more attention to drafts, revision history, research notes and oral explanations. The goal is to determine whether students actually understand the work they submit.

AI detectors can also make mistakes, so a detection score alone may not be enough to establish academic misconduct.

I found it interesting that education is increasingly shifting from simply asking “Who wrote this?” to asking “Does the student understand this?”

What do you think is the fairest way to assess students in the age of AI?

I don’t understand why this post is getting downvoted so heavily. There are use cases where AI is beneficial, but helping kids squander their opportunity to get an education and learn to think for themselves is absolutely not one of them. How to prevent this is definitely a conversation worth having.

Be the change you want to see in the world.

Teachers here have reintroduced books and hand written essays and homework. No more laptops.

You say that like schools were doing that in the first place

That’s actually a really interesting approach. My post wasn’t meant to argue that laptops or AI should simply disappear from schools, though. I was more interested in discussing how AI is changing the way students learn and how teachers are adapting to it.

Exactly. That’s how I see it too. AI can be extremely useful in education, but there’s a big difference between using it as a learning tool and using it to avoid learning altogether. I think that distinction deserves more discussion.

I didn’t mean to suggest that schools have always done this. My point is that this is an ongoing conversation, and AI is making the issue much more visible now. Schools are already having to rethink how students learn, write, research and complete assignments because tools like ChatGPT have changed the situation.

Absolutely. And I think part of that change is having honest conversations about how we use new technology, especially when education is involved.

I found it interesting that education is increasingly shifting from simply asking “Who wrote this?” to asking “Does the student understand this?”

I don't think this framing is quite accurate. Before AI, the final paper's quality was the metric for "did the student understand this". With AI tools, that metric disappeared, and educators are now looking for a replacement heuristic.

Exactly. That’s what I find most interesting about this shift. AI is forcing educators to look beyond the final piece of writing and think more about whether the student actually understands what they’re presenting. I think that’s a much bigger change than simply asking whether AI was used.

It was never a good metric, AI just made it more obvious and worse.

As true as that may be, AI sure isn’t helping.