Fair and Transparent Standards for AI-Related Academic Misconduct

The rapid growth of artificial intelligence has created a legitimate challenge for colleges and universities seeking to protect academic integrity. But, it has created an equally important challenge: how do universities reliably determine whether a student actually used AI?

Recent events at LSU illustrate why this question deserves broader attention. LSU has experienced a substantial increase in AI-related academic misconduct cases and has expanded staffing and case-management resources to address the growing volume. While additional resources may help cases move more quickly, increased capacity is not the same as process improvement. The larger question is how universities measure whether their AI-related determinations are accurate, consistent, and fair—not simply how efficiently cases are processed.

AI-detection tools are not infallible, and even their developers caution against using detector results as conclusive proof of misconduct. Yet students accused of AI use can face serious consequences involving grades, scholarships, honors eligibility, academic progress, graduate or professional education, reputations, future opportunities, and mental well-being. When the consequences are this significant, the process used to reach the determination must be worthy of the same confidence universities expect students to place in their academic institutions.

Recent LSU cases and public reporting have raised broader questions about how suspected AI use is evaluated: What evidence must be independently verified before an allegation proceeds? Are case managers applying consistent standards? Are students clearly shown the specific passages, citations, sources, or other evidence supporting the allegation? Are students clearly informed of their hearing and appeal rights? What happens when contrary evidence emerges after a determination? And does the university track erroneous, dismissed, reversed, or unsupported allegations and use those findings for process improvement and staff training?

We are asking public colleges and universities to adopt clear and consistent safeguards:

• Independent Human Verification: AI-detection results should inform—not replace—human judgment. Material evidence should be independently evaluated before a misconduct determination is made.
• Specific Notice and Evidence: Students should be clearly informed of the specific passages, citations, sources, or other evidence supporting an allegation and be given a meaningful opportunity to respond.
• Consistent Standards: Faculty, case managers, investigators, and hearing officers should follow documented and consistently applied standards for evaluating suspected AI use.
• Meaningful Review and Correction: Students should have clearly explained hearing and appeal rights, including a process for considering material contrary evidence. Institutions should have procedures for correcting academic, financial, and disciplinary consequences when allegations are determined to be unsupported.

Fair and Transparent Standards for AI-Related Academic Misconduct

The rapid growth of artificial intelligence has created a legitimate challenge for colleges and universities seeking to protect academic integrity. But it has created an equally important challenge: how do universities reliably determine whether a student actually used AI?

Recent events at LSU illustrate why this question deserves broader attention. LSU has experienced a substantial increase in AI-related academic misconduct cases and has expanded staffing and case-management resources to address the growing volume. While additional resources may help cases move more quickly, increased capacity is not the same as process improvement. The larger question is how universities measure whether their AI-related determinations are accurate, consistent, and fair—not simply how efficiently cases are processed. LSU expanded resources to keep up with AI cheating cases, provost says

AI-detection tools are not infallible, and even their developers caution against using detector results as conclusive proof of misconduct.  Turnitin — Using the AI Writing Report Yet students accused of AI use can face serious consequences involving grades, scholarships, honors eligibility, academic progress, graduate or professional education, reputations, future opportunities, and mental well-being. When the consequences are this significant, the process used to reach the determination must be worthy of the same confidence universities expect students to place in their academic institutions.

Recent events at LSU and public reporting raise broader questions that every public college or university using AI-detection evidence should be able to answer:

• Accuracy and Human Verification: How does the institution determine whether an AI-related allegation is accurate, and what independent human verification is required before an allegation or finding is made?
• Consistent Decision-Making: Are faculty, case managers, investigators, and hearing officers applying documented and consistent evidentiary standards, and how does the institution measure consistency among decision-makers?
• Notice and Evidence: Are students clearly told what specific passages, citations, sources, or other evidence support the allegation so they can meaningfully respond?
• Student Rights: Are students clearly informed of their hearing and appeal rights, how to exercise those rights, and what options exist when new or contrary evidence becomes available?

• Correcting Errors: What happens when an AI-related allegation is later found to be erroneous, unsupported, dismissed, or reversed? Are grades, academic and disciplinary records, scholarships, honors eligibility, and other consequences promptly and fully corrected, and what steps are taken to remedy harm that may already have occurred because of the allegation or delay in resolving it?

• Case Prioritization and Time-Sensitive Student Impact: Does the institution use any form of prioritization, ranking, or triage when assigning and reviewing AI-related academic misconduct cases? If so, are cases prioritized when delays could affect a student's scholarship eligibility, financial aid, ability to register or schedule required courses, graduation timeline, honors eligibility, admission to graduate or professional programs, or other time-sensitive academic opportunities?

  • If no prioritization process exists, how does the institution prevent administrative delays from creating consequences that may be difficult or impossible to reverse even if the student is ultimately cleared?

• Accountability and Process Improvement: Does the institution track unsupported allegations, dismissals, reversals, appeals, and other outcomes, and use that information for root-cause analysis, process improvement, policy changes, and continuing faculty and case-manager training?

• Measuring Success and Student Impact: How does the institution measure whether its AI-related academic misconduct process is actually working—not merely by the number of cases processed or the speed at which they are resolved, but by their accuracy, consistency, fairness, and impact on student academic, financial, reputational, and mental well-being?

We are asking public colleges and universities to adopt clear and consistent safeguards:

• Independent Human Verification: AI-detection results should inform—not replace—human judgment. Material evidence should be independently evaluated before a misconduct determination is made.
• Specific Notice and Evidence: Students should be clearly informed in writing of the specific passages, citations, sources, or other evidence supporting an allegation and be given a meaningful opportunity to respond.
• Consistent Standards: Faculty, case managers, investigators, and hearing officers should follow documented and consistently applied standards for evaluating suspected AI use.
• Meaningful Review and Correction: Students should have clearly explained hearing and appeal rights, including a process for considering material contrary evidence. Institutions should have procedures for correcting academic, financial, and disciplinary consequences when allegations are determined to be unsupported.
• Measure What Matters: Universities should track not only case volume and processing times, but also dismissals, reversals, unsupported allegations, consistency among decision-makers, student-welfare impacts, and other measures necessary to determine whether the process is actually working. Those findings should be used for root-cause analysis, process improvement, and continuing training.

This petition is not about excusing academic cheating or overturning any individual student's case. Universities have both the right and responsibility to protect academic integrity. But, protecting academic integrity also requires protecting the integrity of the process used to enforce it.

Students who misuse AI should be held accountable. Students who do not should have confidence that the process is capable of recognizing the difference.

We ask higher-education leaders, governing boards, policymakers, and other stakeholders to establish transparent, evidence-based standards for AI-related academic misconduct at public colleges and universities.

Please sign this petition if you believe academic integrity requires integrity in the process used to enforce it.