Most Colleges Have No AI Policy for Admissions. Here's What That Means for You.

The Numbers Are Striking -- and They Agree

In September 2025, Kaplan surveyed 220 college admissions officers about their stance on AI in application essays. The headline finding: 68% of admissions offices have no formal AI policy. Only 2% explicitly allow AI use. And 30% now ban AI outright, up from 25% the year before.

Our own data -- drawn from a review of 150+ university admissions websites -- tells the same story. 67% of schools (111 out of 166) are classified L0 under our L/D/E framework, meaning they have no admissions-specific statement on AI. Our L1 category (schools that explicitly permit AI) accounts for 8% of schools (13 out of 166), which tracks with Kaplan's finding that only a small fraction have actively embraced AI.

Two completely independent datasets. Nearly identical results. The conclusion is clear: the majority of colleges have said nothing specific about AI in admissions essays.

This post examines what that silence means for applicants, why L0 is not the same as "safe," and what you should do if you are applying to schools that have not published a policy.

The L0 Problem: What "No Policy" Actually Means

When a school is classified L0, it means we reviewed their admissions website, application portal, blog posts, and FAQs and found no language specifically addressing AI tools in admissions materials. The school may have a general honesty pledge. It may have an academic integrity policy for enrolled students. But it has not told applicants whether AI is allowed, restricted, or banned in essays.

This is not the same as permission. It is ambiguity.

Consider four of the most selective schools in the country:

• Harvard University (L0/D0/E1): No explicit AI admissions policy. Harvard's admissions office has published no guidance on AI in application essays. They use manual review and may look for voice inconsistencies, but there is no stated rule to follow.

• MIT (L0/D0/E0): No policy at the institutional level. MIT's Chemistry graduate program does allow AI tools (L1 at the program level), but the university as a whole has no admissions AI guidance and no stated enforcement.

• Stanford University (L0/D1/E1): Stanford encourages applicants to let their "genuine voice come through" and optionally note AI assistance, but has no formal ban or permission statement. The Graduate School of Business, however, explicitly prohibits AI (L4/D0/E3) -- a dramatically different posture within the same university.

• Duke University (L2/D0/E1): Duke is actually not L0 -- it is one of the schools that has addressed AI directly. Dean Guttentag stated that Duke no longer scores essays numerically because of AI concerns. But even Duke's institutional stance allows line-level editing, while Duke Law bans AI entirely (L4/D0/E2).

The first three schools -- Harvard, MIT, and Stanford -- are all L0 at the institutional level. They are among the most sought-after universities in the world, and they have published no specific guidance on whether you can use ChatGPT to edit your personal statement.

L0 Does Not Mean Safe

Here is the critical point most applicants miss: even at schools with no AI admissions policy, the Common Application's fraud policy still applies.

The Common App's terms of use require applicants to certify that their application materials are their own work. Submitting AI-generated content as your own is treated as a form of misrepresentation. This is not a hypothetical -- it is the binding agreement you sign when you submit through the Common App, and it covers over 1,000 member institutions.

So a school can be L0 -- no stated AI policy -- while the application platform itself treats AI-generated essays as fraud. You are not in a policy vacuum. You are in a space where the platform-level rule is strict even when the school-level rule is absent.

Beyond the Common App, consider the practical risk: admissions readers have been reading thousands of essays per cycle for years. They are attuned to voice, specificity, and the kind of personal detail that AI tools struggle to produce. An AI-polished essay that reads like every other AI-polished essay will stand out for the wrong reasons -- not because of a detector flag, but because of a human reader's intuition.

The Trend: More Policies Are Coming

Kaplan's data shows movement. In 2024, 25% of admissions offices reported banning AI. By 2025, that number rose to 30%. The direction is clear: schools that currently have no policy are watching what early movers do, and many will eventually publish their own rules.

Our data breaks down the 55 schools (33%) that are not L0 as follows:

• L1 (Permissive): 13 schools (8%) -- including Dartmouth, Rice, Penn, and NYU
• L2 (Line-Level Editing): 26 schools (16%) -- including Georgia Tech, Duke, and Cornell
• L3 (Brainstorming Only): 7 schools (4%) -- including Yale and UC Davis
• L4 (Prohibited): 9 schools (5%) -- including Georgetown, Brown, and BYU

The schools with explicit policies span the full spectrum. There is no consensus emerging toward permissiveness or prohibition. What is emerging is a consensus that schools should say something.

For applicants, this means the policy landscape you encounter in 2026-2027 may look different from what you see today. A school that is L0 this cycle could publish an L3 or L4 policy next cycle.

Practical Advice for Applicants at L0 Schools

If you are applying to schools that have no AI admissions policy, here is what we recommend:

1. Treat L0 as "cautious" rather than "permissive." The absence of a policy does not mean anything goes. Default to the most conservative interpretation: use AI for brainstorming and basic proofreading, but write and revise in your own words.

2. Check program-specific policies. Even if the institution is L0, the specific program you are applying to may have its own rules. Stanford's GSB (L4), MIT's Chemistry program (L1), and Duke Law (L4) all differ from their institution-level classifications. Always check the program's own admissions page.

3. Respect the Common App's terms. If you are applying through the Common App, you are bound by its fraud policy regardless of the individual school's stance. Your essays must be your own work.

4. Write something only you could write. The best protection against both AI detection tools and human reader skepticism is specificity. Name the lab you worked in. Describe the exact moment you changed your mind about your major. Reference the conversation with a specific professor. AI cannot fabricate the details of your life.

5. Keep records. Save your drafts, outlines, and revision history. If a school ever questions the authenticity of your essays, having a clear paper trail of your writing process is your strongest defense.

The Disclosure Question: Should You Volunteer?

Even at D0 schools -- schools that do not ask about AI use -- some applicants wonder whether they should proactively disclose that they used AI tools for brainstorming or editing.

Our take: if you used AI as a minor tool (grammar checking, brainstorming prompts, getting unstuck on an outline), there is no obligation to disclose at a D0 school, and doing so is unlikely to help your application. The essay should speak for itself.

If you used AI more substantially -- generating drafts, paraphrasing paragraphs, structuring arguments -- you should reconsider whether the essay is truly your own work, regardless of disclosure.

The question is not "should I tell them?" The question is "is this still my essay?" If the answer is yes, write it well and submit it with confidence. If the answer is uncertain, revise until the answer is yes.

For schools that do require disclosure, our AI disclosure tool can help you draft an honest, well-framed statement about your AI use.

Explore the Data

• Browse all 150+ schools -- filter by L0 to see every school with no explicit policy.
• AI disclosure tool -- draft a disclosure statement if your school requires one.
• Full methodology -- understand how L/D/E classifications are determined.

---

Related Reading

• Do Top 10 Colleges Check for AI? Official Policies
• Do Colleges Use AI Detectors? The Truth About Turnitin
• What AI Detection Tools Are Colleges Actually Spending On?
• Which Colleges Use AI to Read Your Essays?
• How We Classified 150+ University AI Policies