In 2023, the University of California system did not adopt AI detection tools across its campuses. Instead, several UC teaching and learning centers publicly cautioned faculty against relying on tools like Turnitin's AI indicator or GPTZero to prove misconduct. If you have been accused at a UC campus based on a detector score, that institutional posture is directly relevant to your response.
What the UC system actually said in 2023
The UC system does not centrally mandate a single AI detection tool. In 2023, guidance emerged instead through campus-level teaching and learning centers, academic senate committees, and instructional technology offices. The consistent thread across campuses like UC Berkeley, UCLA, UC Davis, and UC San Diego was caution: AI detectors were described as unreliable enough that a score alone should not form the basis of an academic misconduct allegation.
UC Berkeley's Center for Teaching and Learning published guidance in 2023 advising instructors that AI detectors produce false positives and false negatives, and that they should not be used as the sole basis for a suspicion of misconduct. UCLA similarly published faculty guidance flagging accuracy concerns. Other campuses, including UC Davis and UC San Diego, echoed the same reasoning through their teaching centers and student conduct offices.
None of this guidance carries the force of a systemwide policy. But it establishes a documented institutional position that the tools are not reliable enough to be treated as evidence on their own. For students facing accusations at a UC campus, that matters. See our UC Davis case pattern breakdown for how this plays out in practice.
Why the UC system took this position
The 2023 guidance did not emerge in isolation. It tracked closely with peer-reviewed research published the same year showing that AI detectors misclassify human writing at rates too high for evidentiary use.
The most-cited paper is Liang et al. (2023), published in the Cell Press journal Patterns, which found that GPT detectors misclassified TOEFL essays written by non-native English speakers as AI-generated at high rates while performing much better on native English writing. A parallel evaluation by Weber-Wulff et al. (2023), published in the International Journal of Educational Integrity, tested fourteen detection tools and concluded that none performed consistently well enough for institutional decision-making.
Vanderbilt University reached a similar conclusion in August 2023, publicly disabling Turnitin's AI detection feature over accuracy concerns. UC campuses did not disable Turnitin, but their guidance reflected the same underlying research. Our coverage of the Vanderbilt decision covers what the university actually said and how to cite it.
What UC academic conduct policy actually requires
Each UC campus operates under its own Code of Student Conduct, applied through offices like Berkeley's Center for Student Conduct, UCLA's Office of Student Conduct, UC Davis's Office of Student Support and Judicial Affairs (OSSJA), and equivalents at other campuses. The systemwide framework is set out in UC Policy 100.00 (PACAOS), which establishes the baseline rights and procedures.
Two features of UC procedure matter for AI detection cases:
- Standard of evidence. UC campuses generally use a preponderance-of-the-evidence standard. That means the finding must be more likely than not, based on evidence in the record. A probabilistic detector score, on its own, is thin support for that finding when independent research has shown the tools misclassify human writing.
- Right to review the evidence. Students are entitled to know what evidence is being used against them. That includes the specific detector, the specific score, and any human review conducted before the allegation was filed.
Read your campus's conduct code and syllabus carefully. Course-level rules on AI use vary by instructor, and the boundary you are alleged to have crossed has to be one that was actually communicated to you.
How to use UC guidance in your response
If you are responding to an accusation at a UC campus, the 2023 guidance is one of several strands you can weave together. It works best when combined with the underlying research and your own process evidence.
| Evidence type | What it establishes |
|---|---|
| Your campus's teaching-center guidance | The institution has already acknowledged AI detectors are unreliable |
| Liang et al. (2023) and Weber-Wulff et al. (2023) | Peer-reviewed evidence that detectors misclassify human writing |
| Vanderbilt's August 2023 decision | A peer institution disabled the tool over accuracy concerns |
| Your drafts, version history, and research notes | The writing process and authorship are documented |
Your response should state, in writing, that you are requesting the specific detector used, the exact score, and any human review conducted. It should cite the UC guidance from your own campus by name. It should point to the standard of evidence in the applicable conduct code and argue that a detector score, standing alone, does not meet it.
What the 2023 guidance does not do
It does not prohibit instructors from using AI detectors. It does not automatically invalidate an accusation. It does not remove your obligation to respond substantively. And it does not answer the question of whether you actually did what you are accused of doing, which the conduct process will still test.
What it does is give you a documented institutional acknowledgment that the tool at the center of many of these accusations is not reliable enough to stand on its own. That is a foundation to build on, not a case that argues itself. If you are preparing a written response, NotBot generates a personalized defense package that cites your campus's guidance, the peer-reviewed research, and your specific writing process, ready in about a minute. For post-finding cases, the appeal package covers the procedural grounds that matter at that stage, and the procedural rights FAQ covers what you are entitled to request from your campus conduct office.
If your case carries suspension, expulsion, or visa consequences, consult an education law attorney before your hearing. The research and the UC guidance are useful, but they are not a substitute for legal advice when the stakes are that high.
Build your UC defense package
A personalized response that cites your campus guidance, the research, and your writing process.
Get your defense package$49 one-time · Generated in 60 seconds