PA School AI Policies 2026 — Why 18 of 20 Programs Are Silent

PA School AI Policies 2026 — Why 18 of 20 Programs Are Silent (And What Actually Binds You)

Short answer: We surveyed 20 of the most prominent PA programs in the United States looking for program-specific AI policies for the 2026-2027 application cycle. Only one — University of Washington MEDEX Northwest — publishes its own admissions-page AI policy that goes beyond CASPA's central rule. The other 18 are silent at the program level. The silence is not an oversight. It's a deliberate policy choice: CASPA itself has the strictest anti-AI certification language in any centralized health-professions application, and individual programs have decided that one binding rule is enough.

This article is the field guide. We name the 20 programs we surveyed, show what each one actually publishes (or doesn't), and explain what binds you regardless of whether your specific program has its own page about AI.

TL;DR: the binding rule for almost every PA applicant in 2026

If you are applying to PA school in the 2026-2027 cycle, the rule that actually governs your essay is the CASPA Applicant Certification, which you must sign before submission. It is the strictest published AI prohibition in centralized US health-professions admissions:

"I certify that all written passages within my CASPA application, including but not limited to, personal statements, essays, and descriptions of work and education activities and events, are my own work, and have not been written, in whole or part, by any other person or any generative artificial intelligence platform, technology, system, or process, including but not limited to ChatGPT (collectively, 'Generative AI'). I am strictly prohibited from using Generative AI to create, write and/or modify any content, in whole or part, submitted in CASPA and/or provided to PA programs on behalf through any means of communication."

— PAEA 2025-26 CASPA Policies & Procedures, carried forward into the 2026-27 PAEA Admissions Suite of Products P&P document.

We have a clause-by-clause decode of the CASPA AI certification statement for the linguistic deep dive. The short version: it bans not just writing, but also modification, of any content. There is no allowance for brainstorming, editing, or rephrasing — the kind of permissions AMCAS explicitly grants pre-meds.

Why this matters: PA admissions is the most centralized in health professions

Of the four major US health-professions central application services, CASPA is the most strict and the most centralized:

We covered this comparison in detail in Can You Use ChatGPT for Your Medical School Application? AMCAS, AACOMAS, CASPA, TMDSAS Compared. The takeaway for PA applicants specifically: you are operating under the strictest centralized rule of any health-professions applicant in the US. Your friends applying to MD school can use ChatGPT to brainstorm and edit. You cannot.

This explains the silence at the individual program level. CASPA has already done the policy work. Most PA programs see no need to publish an additional rule that would either be redundant or contradict the central one.

Methodology

We surveyed 20 PA programs in April 2026. The list below is not exhaustive — there are roughly 300 ARC-PA-accredited PA programs in the US — but it covers the highest-prestige, highest-volume programs across geographies, university affiliations, and institutional types.

For each program we searched:

1. The PA program's own admissions page
2. The university's general AI or academic integrity policy
3. The PA program's FAQ
4. Recent press releases or interviews where the program discussed AI

We were looking specifically for program-specific AI policies that go beyond simply mirroring CASPA's central rule. Generic university-wide academic integrity language that applies to enrolled students (not applicants) was noted but did not count.

The full underlying research, with verbatim quotes and source URLs for each program, is the source for this summary. We will publish a structured /ai-policies/pa-schools aggregation page following the model of our existing /ai-policies/medical-schools hub once the dataset is verified end-to-end.

The 20 programs surveyed

Sorted by stance.

Programs with a published, program-specific AI policy (1)

1. University of Washington MEDEX Northwest

Status: Has a published policy. Stricter than CASPA. Reserves the right to use AI detection tools.

MEDEX Northwest is the only program in our 20-program survey to publish a program-specific AI policy on its admissions page that adds nuance beyond CASPA. The policy explicitly permits limited AI use for non-substantive editing (spelling and grammar) but bans substantive content generation, and notably reserves the right to use AI detection tools — even though PAEA itself has cautioned member programs that detection is unreliable. We have a dedicated deep dive on the MEDEX Northwest AI policy because it is genuinely unique in the landscape.

Source: MEDEX Northwest applicants page.

Programs with informal advisory language (1)

2. University of Iowa (pre-health advising, not the PA program admissions office)

Status: Informal advisory only. The PA program itself is silent.

The University of Iowa Academic Advising Center publishes pre-health personal-statement tips that advise pre-health students against using ChatGPT. The wording, per indexed source material:

"Don't use generative AI (Chat GPT). Admissions committees will be able to tell."

This is advising (a service provided by the university to its students), not formal admissions policy. The Iowa PA program's actual admissions page does not surface an AI policy. CASPA's central prohibition still binds Iowa PA applicants. We are flagging this as a half-policy because it shows that at least one major university is publishing AI guidance to its pre-PA students even when the program itself stays silent.

Source: Iowa Academic Advising Center pre-health personal statement tips.

Programs that are silent at the program level (18)

The following 18 programs have no AI policy published on their PA admissions pages as of April 2026. Each defers entirely to the CASPA central certification.

3. Duke University PA Program (Duke School of Medicine). Duke's admissions leadership has commented publicly on AI in the Admissions Straight Talk podcast Episode 591 — Dr. April Stouder, the Associate Program Director and Director of Admissions, told the host that "AI-generated emails and essays just don't ring very true." That commentary, however, has not been added to the program's admissions page.

4. Yale School of Medicine — Physician Associate Program (Yale Medicine PA). Notably, Yale College's undergraduate admissions AI policy is one of the most explicit anti-AI statements in higher education — but it formally applies only to undergraduate admissions, not to the PA program. The Yale PA Online Program is winding down (final class graduates 2026, no new students admitted). Yale's residential PA program does have unusually demanding supplemental essay prompts of its own — see our prompt-by-prompt breakdown of the Yale PA program supplemental essays decoded.

5. Emory University PA Program (Emory School of Medicine). Requires a supplemental narrative essay. No published AI policy alongside it.

6. George Washington University PA Program (GW SMHS). GW's general academic integrity code covers admissions falsification but does not name AI. Defers to CASPA.

7. Wake Forest University PA Program (Wake Forest School of Medicine). Holistic-review supplemental application. No AI policy on the program page.

8. Northeastern University PA Program (Bouvé College of Health Sciences). No program-level AI policy.

9. Pace University PA Programs (both Pleasantville and NYC/Lenox Hill). Pleasantville admissions. No program-level AI policy. Both programs include personal-statement components.

10. Quinnipiac University PA Program (Quinnipiac MHS PA). Quinnipiac has a comprehensive university-wide Generative AI policy that applies to enrolled students and faculty — not admissions. PA applicants are bound by CASPA.

11. University of Iowa PA Program (Iowa Carver College of Medicine). Pre-health advising office offers informal guidance (above), but the program admissions page is silent.

12. Oregon Health & Science University (OHSU) PA Program (OHSU School of Medicine). OHSU PA does not have a separate supplemental application — applicants answer OHSU-specific questions inside the OHSU portal of CASPA, and those answers fall under the central CASPA prohibition.

13. Stanford University Master of Science in PA Studies (Stanford School of Medicine). Stanford Graduate Admissions has general advisory language ("Think very carefully about the use of generative AI bots, as these may lead to statements that are not authentic to your own experiences") but no hard prohibition. Stanford PA defers to CASPA.

14. University of Wisconsin-Madison PA Program (UW-Madison School of Medicine and Public Health). UW-Madison undergraduate admissions has an unusually permissive policy ("We will not disqualify an applicant found to have used or suspected of using AI in their admissions essays") — but that policy formally applies only to undergraduate admissions, not the PA program.

15. University of Florida PA Program (UF College of Medicine School of Physician Assistant Studies). No program-specific AI policy.

16. University of Colorado CHA/PA Program (CU Anschutz). Requires both CASPA and a CU CHA/PA supplemental application. Neither surfaces an AI policy.

17. Methodist University PA Program (Methodist University). Methodist's personal statement guidance page is one of the most cited authoritative sources in PA admissions writing — it predates the AI debate and addresses ghost-writing concerns by emphasizing voice and lived experience. It does not name AI. Defers to CASPA.

18. University of Utah PA Program (UPAP) (Utah School of Medicine). Utah launched ChatGPT Edu university-wide for enrolled students in 2025-26, but this does not govern admissions. UPAP defers to CASPA.

19. Stony Brook University PA Program (Stony Brook ELPA). Stony Brook's first-year undergraduate admissions has an explicit ban on AI-generated writing samples. That policy formally applies only to first-year undergraduate admissions, not PA.

20. Drexel University PA Program (Drexel CNHP). Drexel signed an enterprise ChatGPT Edu license in December 2025 — broad institutional AI adoption — but admissions pages do not name AI. Defers to CASPA.

21. USC Keck PA Program (MPAP) (USC Keck). USC adopted ChatGPT Edu university-wide in December 2025. The USC Office of Academic Integrity has a generative AI policy that applies to enrolled students. Neither governs PA admissions. Defers to CASPA.

(Yes, that's 21 program names — Pace counts as two separate program campuses, both surveyed.)

What the silence actually means for applicants

This pattern — CASPA at the center, programs silent at the edges — has three practical implications.

1. Stop searching for your specific program's AI policy

If you Google "[Your University] PA program AI policy", you will almost certainly find nothing program-specific, or you will find a university-wide policy that does not actually apply to admissions (the Yale undergraduate policy, the Stony Brook first-year undergraduate ban, the UW-Madison undergraduate permissiveness). None of those bind you. The CASPA certification does. Read it once, understand it, and move on.

2. The "but my program said X" defense doesn't exist

Some applicants assume that because their target program has not specifically banned AI use, they have wiggle room. They do not. The CASPA certification is signed at the application level, not the program level. You sign once and it covers every program you apply to. There is no individual-program override that loosens the rule. (There may be one that tightens it — that's MEDEX Northwest, where the program publishes additional restrictions on top of CASPA. We have not found a program that has published looser rules.)

3. Detection is contested — but stricter at MEDEX

PAEA itself has been remarkably honest about the unreliability of current AI-detection tools. Per their published guidance for PA programs (What Your Program Should Know About AI and Admissions), PAEA will not initiate a CASPA investigation where the sole basis for suspicion is AI-detection software. They cite false-positive and false-negative concerns and recommend in-person essay-writing during interviews as a verification mechanism instead.

But individual programs may go further. MEDEX Northwest is the one program in our survey that explicitly says it will run AI detection tools and use AI-supported systems during admissions review — putting it slightly out of step with PAEA central guidance. If MEDEX is on your list, treat the policy as binding even though most other programs are silent.

For more on the false positive problem in AI detection, see our piece on flagxiety — the term we coined for the anxiety students feel about being flagged by AI detectors even when they wrote every word themselves. PAEA-funded research on how detectors actually perform on PA application essays — including a measured false-positive rate for human writing — is summarized in our PAEA AI detection research breakdown. And for an always-current list of every PA program's published AI position in one place, see our PA program AI policies aggregator for 2026.

Special note: the new 2026-2027 CASPA AI essay

Beginning with the 2026-2027 cycle, CASPA has added a new "Situational Decision-Making Question" (also referred to as the AI and Technology essay), replacing the COVID-19 Impact Essay. The new prompt asks how future PAs should learn to use emerging technologies like AI, telemedicine, and wearable health devices thoughtfully — specifically including environments where access to technology may be limited.

This is the most awkward situation in the entire CASPA application. You have to write thoughtfully about AI as a clinical tool while CASPA simultaneously prohibits using AI to draft, edit, or rephrase that essay. The discipline of writing about AI without AI is part of the test.

We have a complete guide to the new CASPA AI and Technology essay, including the verbatim prompt, the hidden "limited access" hook, 7 worked angles, and 5 mistakes applicants will make. If you are writing this essay for the 2026-2027 cycle, start there.

What we'd add to this picture if we could

Three things this survey could not capture and that future versions of this article will improve:

1. Smaller programs. Our 20-program list skews to large, prestigious, well-known programs. There are roughly 300 ARC-PA accredited PA programs in the US. Smaller programs may have published AI policies that we did not surface in this round of research.
2. Direct-fetch verification of program pages. A few of the program pages we cite were indexed via search snippets rather than direct page fetches due to a tooling limitation during research. We will refresh and verify each citation directly before any production aggregation page goes live at /ai-policies/pa-schools.
3. Policy version history. PA programs rarely publish version dates or change logs on policy pages. A program that is silent today may publish something tomorrow. We expect more programs to add explicit AI policies during the 2026-2027 cycle, especially after the new AI essay rolls out and admissions readers see how applicants engage with it.

If you know of a PA program with a published, admissions-page AI policy that is not on our list, the best place to send it is our medical content team via the GradPilot contact form. We will add it to the next revision.

---

Related Reading

Medical school essays hub: Medical School Essays — The Complete Guide to AMCAS, AACOMAS, CASPA & TMDSAS — every medical school essay guide on the site, organized by application system, topic, and applicant profile.

The CASPA + AI policy cluster:

• CASPA AI and Technology Essay 2026-2027 — Prompt + 7 Angles — the new CASPA essay this article references throughout
• CASPA AI Certification Decoded — What It Actually Bans — clause-by-clause read of the certification statement
• MEDEX Northwest AI Policy — The One PA Program with Rules — the only program with its own published policy
• Can You Use ChatGPT for Your Medical School Application? AMCAS, AACOMAS, CASPA, TMDSAS Compared — the four-system comparison
• What Is Flagxiety? The AI Detection Anxiety Reshaping How Students Write — false positives and the detection landscape
• Medical School AI Policies 2026: AMCAS Rules & 60+ Schools — our AI policy hub for medical admissions

The CASPA writing cluster:

• The CASPA Life Experiences Essay: What the New Prompt Actually Asks
• Sample CASPA Essay Analysis: How 40+ Applicants Got Into Top PA Programs
• CASPA Personal Statement Topics to Avoid