New York University's Tandon School of Engineering has released a comprehensive guide for managing artificial intelligence in academic settings, addressing a growing challenge for institutions struggling to maintain academic integrity as generative AI tools proliferate in classrooms.
The 18-author framework, titled "Protecting Learning: A Practical Guide to AI-Use Policy for Instructors, Departments, and Schools," emerged from faculty collaboration within Tandon's AI Across the Curriculum initiative. According to AI Weekly, the guide was published in November 2026 and remains open for public feedback through the end of the month, with revisions planned for early 2027.
Detecting Suspicious Patterns
The timing of the release reflects mounting institutional concerns about how students deploy AI systems in ways that undermine learning. Tandon researchers pointed to a concrete case study: one foundational course where students completed homework assignments with near-perfect scores, yet averaged roughly 50% on proctored examinations. That disparity signals potential AI assistance on unsupervised work, the school argues, and represents precisely the kind of anomaly educators should monitor and investigate.
The disconnect between unproctored and proctored assessment performance offers institutions a practical detection mechanism that requires no sophisticated technological infrastructure. Educators can apply this principle across disciplines and course formats, making it accessible to schools with limited resources for AI monitoring systems.
Practical Framework for Educators
Rather than prescribing one-size-fits-all restrictions, the guide takes a nuanced approach to AI governance. It acknowledges that artificial intelligence can serve legitimate pedagogical purposes while establishing safeguards against academic dishonesty. The framework provides institutions with guidance for developing policies tailored to their specific educational missions and student populations.
- Analyzing performance variations between assignment and exam contexts
- Establishing clear institutional expectations around AI tool usage
- Designing assessments resistant to AI completion without human engagement
- Training faculty on emerging AI capabilities and limitations
- Building departmental consensus on enforcement mechanisms
Wider Industry Implications
The release arrives as universities nationwide grapple with policies around generative AI. Schools have adopted divergent approaches, ranging from blanket prohibitions to regulated integration models. Tandon's research-backed guide offers evidence-based strategies that could influence policy decisions across higher education.
The collaborative authorship and extended public comment period suggest the institution views this as a living document rather than final pronouncement. Revisions will likely incorporate feedback from educators across different disciplines and institutional types, potentially creating broader applicability.
As generative AI capabilities continue advancing, the gap between what students can accomplish with AI assistance and what they can demonstrate under controlled conditions will likely become an increasingly important signal for academic integrity initiatives. Universities that develop sophisticated understanding of these patterns may gain competitive advantages in maintaining rigorous standards while responsibly integrating powerful new technologies into curriculum.



