Princeton University just killed a 133-year-old tradition because of ChatGPT.

The university is abandoning its legendary honor code and introducing proctored exams for the first time since 1893. Students will no longer be trusted to take tests unsupervised. The reason: AI cheating tools have made the self-policing system impossible to maintain.

Let that sink in. One of America's most prestigious universities, with a founding principle built on academic integrity and student trust, just admitted that generative AI broke their system in less than two years.

Princeton's honor code wasn't just a policy, it was part of the school's identity. Students signed a pledge. They took exams alone, without proctors. If they saw cheating, they were honor-bound to report it. For over a century, it worked.

Then came GPT-4, Claude, and every other model that can write coherent essays in seconds. Suddenly, the honor system had a massive, invisible hole. Professors couldn't tell which work was student-written and which was AI-generated. Students who were honest felt like they were competing against machines. The trust collapsed.

This is bigger than one school's policy change. If AI can force a 133-year institutional tradition to collapse in a single academic year, what other trust systems are next?

On Reddit, commenters are already asking the obvious questions. What happens to professional licensing exams? Bar exams? Medical boards? All of them currently assume the test-taker is working alone. All of them assume you can verify independent knowledge. AI just made that assumption obsolete.

Some universities are trying to adapt. They're requiring handwritten exams. Oral defenses. Project-based assessments that are harder to fake. But those approaches don't scale. And they don't solve the fundamental problem: we no longer have a reliable way to verify what someone knows versus what an AI told them.

I've talked to educators who are genuinely struggling with this. The technology moved faster than their ability to redesign curricula. Some are angry. Some are resigned. Most are just confused about what "learning" even means when students have instant access to superintelligent tutors.

Princeton isn't overreacting. They're recognizing reality. The honor code worked when cheating required effort and risk. AI removed both. You can cheat from your dorm room, with zero human interaction, and produce work that's indistinguishable from your own.