# AI Surveillance Botched 150,000 Mexican University Exams > Source: > Published: 2026-08-02 18:27:46+00:00 Mexico's most prestigious university handed its admissions process to an artificial intelligence surveillance system, and now roughly 150,000 test-takers are paying the price — with thousands of results annulled and every student left in limbo, including those nobody claims cheated. It's a preview of exactly the kind of algorithmic gatekeeping Big Ed and Big Tech want to normalize in American classrooms. The National Autonomous University of Mexico, known as UNAM, moved its admissions exam online for the first time this year to make the test more accessible. To guard against cheating, the university deployed an AI system that it said would "allow for the capture and recording of any behavior and action carried out by each applicant," supporting human monitors. Instead, the results showed an unusually high number of perfect scores — suggesting students either used AI tools themselves or obtained the questions in advance. UNAM responded by annulling roughly 3,000 of the 150,000 tests administered. But the collateral damage is enormous. The university has suspended enrollment entirely while it investigates, meaning every applicant — the innocent included — now sits in bureaucratic purgatory with the new semester scheduled to start August 10. Daniel Fernando García Velázquez, a student who took the exam four times and traveled 250 miles from Veracruz to Mexico City for each attempt, scored a 94 out of 120 on his final try. That score should have secured him a spot in the Communication and Journalism program at UNAM's Aragón School of Higher Studies. He still doesn't know if he's in. The AI system was supposed to be the safeguard that made online testing possible. Instead, it failed to prevent the very cheating it was built to stop, and the university's response — blanket suspension — punishes honest students alongside suspected cheaters. The uproar has reached Mexican President Claudia Sheinbaum, herself a UNAM physics graduate, who suggested the university bring in the attorney general to sort out the mess. UNAM is no ordinary school. Founded in the 16th century and granted autonomy in 1929, it functions as the primary engine of social mobility in a country where private higher education is prohibitively expensive for most families. More than 100,000 applicants compete each year, and 80 to 90 percent are rejected. The university's chronic shortage of seats already sparks regular protests demanding more classroom space. Adding an AI surveillance layer — then freezing everyone's results when the algorithm couldn't do its job — has turned a broken process into a national scandal. Both KRDO and CNN covered the story with essentially identical reporting, framing it as a scandal of student cheating enabled by technology. What neither outlet pressed hard on is the more uncomfortable question: why did UNAM outsource exam integrity to an AI monitoring system in the first place, and who supplied it? The university has not publicly named the vendor or explained what the system actually detected — or failed to detect — before the perfect-score anomaly forced administrators' hands. The lesson isn't that students cheat. Students have always cheated. The lesson is that when institutions replace human judgment with algorithmic surveillance, the machine doesn't just fail — it fails at scale, and the people with the least power absorb the consequences. What happens when this same logic arrives in American school districts, propped up by the same ed-tech vendors selling the same promises? The students in Mexico just found out.