UC Berkeley Law Bans AI For Most Student Work Summer 2026

UC Berkeley School of Law has enacted one of the strictest artificial intelligence policies in higher education, effectively banning students from using generative AI for most graded work beginning in Summer 2026.

The default policy prohibits AI assistance in conceptualizing, outlining, drafting, revising, translating, or editing any assignment submitted for credit. It also bans the use of AI during exams and prevents students from uploading any course materials, including readings, slides, or lecture recordings, to generative AI systems.

Instructors may allow limited AI use only in specially designed courses focused on AI fluency, but they must provide written notice and require full disclosure of how the tool was used. The policy’s stated goal is to protect the development of core cognitive skills essential for lawyering while ensuring fairness.

With Texas universities already experiencing significant AI-driven grade inflation, the Berkeley Law crackdown is drawing close attention from students and administrators across the state.

Colleges Battle AI Surge and Rising ‘Gradeflation’

This move comes as colleges nationwide grapple with widespread student adoption of AI tools. Gallup’s 2026 State of Higher Education study found that 57% of U.S. college students use artificial intelligence in coursework at least weekly, with about one in five using it daily.

A 2026 HEPI survey of UK undergraduates reported even higher figures, with 95% of students using AI in some form and 94% using generative AI to help with assessed work.

Many other institutions have adopted similar restrictive policies. Harvard, Stanford, and Yale require attribution or prohibit the use of undisclosed AI in essays and exams. Duke treats unauthorized use of AI as cheating under its community standards. The University of Chicago emphasizes the verification of AI-generated content and prohibits the entry of confidential data. Caltech leaves decisions to individual instructors but treats unapproved use as an academic integrity violation. SUNY adopted a systemwide AI policy in May 2026 across its 64 campuses, embedding AI literacy into general education while strengthening academic integrity rules.

The Dallas Express previously reported on a University of California, Berkeley working paper analyzing more than 500,000 grades at a large Texas research university from 2018 to 2025. The study found that in courses vulnerable to AI — such as those heavy in writing and coding — the share of A grades increased by about 13 percentage points, or roughly 30% relative to the 2022 baseline, after ChatGPT’s debut. The rise was more pronounced in classes where homework carried greater weight.

Researchers attributed the shift primarily to students using AI for assignments rather than broad improvements in student learning.

This phenomenon, sometimes called gradeflation, has raised concerns about the long-term value of grades as signals of mastery. Faculty surveys show near-universal concern that AI undermines critical thinking, original writing, and academic integrity, with 92% of faculty concerned about plagiarism or dishonesty facilitated by AI, according to a February 2026 College Board survey of more than 3,000 U.S. college faculty.

Texas Universities Take Varied Approaches to AI in the Classroom

In Texas, here are the current AI policies for several of the top universities:

University of Texas at Austin (UT Austin): Instructors set course-specific policies; unauthorized use may violate academic honesty rules. The university encourages responsible adoption through guiding principles rather than a blanket ban.
Texas A&M University: Decentralized approach — no single systemwide ban. Faculty determine AI use on a per-course basis, with unauthorized use treated as academic misconduct under the Aggie Honor Code. Strong emphasis on transparency, citation, and using AI as a learning assistant.
Rice University: Treats undisclosed use of AI to generate ideas or content as plagiarism under the Honor Code. Professors may set their own course policies.
Texas Tech University: Faculty are encouraged to include clear statements on permitted/prohibited AI use in the syllabus. No rigid university-wide prohibition; decisions left largely to individual instructors.
University of Houston (UH): Supports responsible and innovative use of AI tools. The system provides security and privacy guidelines while developing a more formal policy.

Colleges face a balancing act. Some instructors design courses around AI fluency, while others tighten rules to preserve traditional skills.

As AI tools grow more capable, universities must decide how to prepare students for professions where AI use is increasingly common while ensuring graduates retain essential human skills.