Kazakh Universities Spend Millions on AI Detection While Avoiding Real Reform
Kazakh universities are purchasing expensive AI-detection systems to catch student cheating even as the government officially endorses generative AI as a professional tool. The contradiction reveals institutions unprepared to redesign how they assess learning in an era when students routinely use ChatGPT for coursework and dissertations.
By spring 2026, using generative AI in Kazakh universities had become standard practice. Yet universities continued buying AI-detection software rather than rethinking their educational models. Kazakh National Medical University spent approximately $27,000 on detection systems. Toraighyrov University spent around $19,000 across multiple contracts.
The Ministry of Science and Higher Education rejected outright bans on generative AI in 2024 through the "Inter-University Standard for the Use of AI." Officials said AI tools should be treated as instruments, not threats, as long as students maintain academic integrity and transparency.
Detection Systems Don't Deliver Certainty
The detection software universities purchased operates on probabilistic models-they estimate likelihood rather than prove guilt. Kazakhstan's Ministry of Education acknowledged this limitation, stating such tools cannot serve as definitive proof of misconduct.
Antiplagiat.Kazakhstan introduced the most widely adopted detection algorithm. Universities awarded contracts through single-source procurement procedures, strengthening one supplier's market dominance.
University Rules Haven't Caught Up
Many university policies predate mass AI adoption. Documents from Yessenov University and Narxoz University contain no references to AI, neural networks, or text generation.
Even recently updated policies preserve outdated logic. Al-Farabi Kazakh National University's 2025-2026 academic policy formally permits ChatGPT use but simultaneously requires diploma theses maintain 75% originality. This creates a legal contradiction: students can use AI without violating traditional plagiarism rules because generated text isn't technically copied from someone else.
Experts warn that disciplinary actions based solely on detection systems could trigger legal disputes between students and universities.
The Diploma Thesis Problem
If AI systems produce academic papers faster and often better than average students, then evaluating written work no longer measures real competencies. The traditional diploma thesis is losing relevance.
The International Information Technology University offers one alternative. Students defend fully developed startup projects instead of writing traditional thesis papers. The focus shifts from text origin to the student's ability to explain the work, defend it before a commission, and demonstrate practical skills.
Two Paths Forward
Kazakhstan's universities face a choice. They can create assessment formats where AI use is impossible-oral defenses, practical assignments, live problem-solving. Or they can formally recognize AI as an essential professional tool and redesign curricula accordingly.
The current approach-spending tens of thousands on detection systems while maintaining outdated assessment rules-cannot sustain itself. Universities are in a transitional phase where expensive technical fixes are becoming increasingly ineffective.
For educators, this shift demands action now. Consider exploring AI for Education resources that address assessment redesign rather than detection. Understanding how ChatGPT actually works helps educators design assignments where AI becomes a tool for learning rather than a threat to grades.
Your membership also unlocks:
