ASU's AI Policy Just Changed. Here's Exactly What You Can and Can't Do.
Marcus RodriguezYou've probably heard conflicting things about whether you can use AI at ASU. Some professors say it's fine. Others treat it like you pulled out your phone during an exam. Your friend in engineering uses Claude for every assignment. Your friend in English got a warning for using Grammarly's AI rewrite.
The truth is more nuanced than either side admits, and the rules just changed again for 2026. Here's the complete breakdown so you can stop guessing and start knowing.
The big picture — ASU's official stance on AI
Arizona State University has been one of the most AI-forward universities in the country. They partnered with OpenAI in 2024 to integrate AI into campus systems. They launched AI-powered tutoring. They didn't take the knee-jerk "ban everything" approach that some schools did.
But here's the critical thing most students miss: ASU does not have a single campus-wide AI policy that applies to every course. Instead, the university sets broad guidelines and leaves the specifics to individual instructors.
The university-level position boils down to three points:
- 1. AI use must be disclosed when it's part of submitted work
- 2. Individual instructors set their own AI policies in their course syllabi
- 3. The Academic Integrity Policy (AIP) treats undisclosed AI use the same as any other form of unauthorized assistance — which means the same consequences as plagiarism
That last point is the one that matters. ASU doesn't treat AI use as automatically wrong. They treat undisclosed AI use as wrong. The difference is everything.
The three tiers of AI permission at ASU
After talking to students across every college and reviewing dozens of syllabi from the 2025-2026 academic year, AI policies at ASU fall into three clear categories:
Tier 1: AI Prohibited
Some courses ban all AI assistance. Period. This is more common than you'd think, and it's not just the courses you'd expect.
Where you'll see this:
What "prohibited" actually means: No AI at any stage. Not for brainstorming. Not for outlining. Not for "just checking" your grammar with an AI-powered tool. Not for summarizing the reading before you do the reading yourself. If the syllabus says no AI, it means no AI.
The common mistake: Students think "I'll use AI for the outline and then write it myself, so it's fine." In a Tier 1 course, it's not fine. The outline is part of the assignment. If you wouldn't hand your outline to a friend and ask them to structure your paper, don't hand it to ChatGPT either.
Tier 2: AI as a Tool
This is where the majority of ASU courses fall in 2026, and it's also where the most confusion lives.
The general principle: You can use AI for research, brainstorming, and getting unstuck. The final submitted work must be substantially your own. You must disclose what AI you used and how you used it.
What this looks like in practice:
What crosses the line in Tier 2:
The disclosure requirement is non-negotiable. Even in Tier 2 courses, you need to say how you used AI. Most professors accept a simple note at the end of your submission: "I used ChatGPT to brainstorm my thesis options and Perplexity Academic mode to find three of my sources." That's it. Easy. Protects you completely.
Tier 3: AI Encouraged
Some courses at ASU don't just allow AI — they require it.
Where you'll see this:
What this looks like: The professor might assign you to solve a problem using AI tools, evaluate different AI models' outputs, or build something that integrates AI. In these courses, NOT using AI might actually hurt your grade.
Even in Tier 3, there are still rules. You typically need to document your AI usage, explain your prompting strategy, and demonstrate that you understand what the AI produced. Submitting AI output without analysis is still lazy — it's just lazy in a different way.
What counts as "AI use" (it's more than you think)
This is where students consistently get tripped up. When ASU says "AI use must be disclosed," they're talking about more than just ChatGPT conversations.
Things that count as AI use at ASU:
Things that are generally fine (no disclosure needed):
The gray area is real, and when you're in the gray area, disclose. Over-disclosing never hurts. Under-disclosing can end your semester.
How ASU detects AI-generated work
Let's be honest about this because pretending detection doesn't exist helps nobody.
Turnitin's AI detection is integrated into Canvas, which ASU uses for most course submissions. Turnitin assigns an "AI likelihood" score to submitted work. It's not perfect. False positives happen, especially with ESL (English as a Second Language) students whose writing patterns can look similar to AI output.
What professors actually look for matters more than any software:
- Writing quality that suddenly jumps compared to your previous submissions
- Impossibly clean structure — real student writing has quirks, AI writing is suspiciously organized
- Lack of personal voice, examples, or specificity. AI writes in generalities. You write about the specific thing that happened in your specific life.
- Citations that don't exist. This is the number-one giveaway. ChatGPT invents sources. If a professor checks your citations and finds a paper that was never published, that's game over.
- Work that doesn't match your in-class performance. If you can barely write a paragraph during in-class activities but submit polished 5-page papers, that gap tells a story.
- Identical or near-identical submissions from multiple students who clearly used the same AI prompt
The best detection is a professor who knows your writing. Software catches the lazy attempts. Professors catch the clever ones. And the smaller the class, the more they notice.
What happens if you get caught
ASU's Academic Integrity Process has real consequences. Understanding them is important not to scare you but so you can make informed decisions.
First offense (most common outcome): The instructor reports the violation to the Dean of Students. You'll meet with the Academic Integrity Officer. Typical outcome is a grade penalty on the specific assignment (often a zero) plus an "XE" notation that goes in your file. Many first offenses include an educational component — you might have to complete an academic integrity course.
Second offense: This is where it gets serious. A second violation typically means failing the course. Not just the assignment — the entire course. An "XE" grade (failure due to academic dishonesty) appears on your transcript. Employers and graduate schools can see this.
Serious or repeated violations: Suspension or expulsion from the university. This is rare but real. Students who run paper mills, share AI prompts for coordinated cheating, or repeatedly violate after warnings can face removal.
The permanent record problem: Academic integrity violations at ASU are tracked centrally. Even if your professor handles it informally, there's a record. If you apply to graduate school, professional programs, or certain jobs, this can surface.
The bottom line: it's never worth it. Using AI well is easy. Getting caught using it badly follows you for years.
The smart way to use AI at ASU
Here's the framework that keeps you learning, keeps you compliant, and actually makes AI useful:
1. Read the syllabus AI policy on day one. Not the first week. Day one. Highlight the AI section. If there isn't one, email the professor and ask. This one action prevents 90% of problems.
2. Use AI for understanding, not production. Ask Claude to explain why your code throws an error — then fix it yourself. Ask ChatGPT to break down a complex reading — then write your own analysis. The AI is a tutor, not a ghostwriter.
3. Use tools with built-in citations. Perplexity's Academic mode and Lazy AI both provide inline source citations. This means when you research with AI, you get real sources you can verify and cite properly. No hallucinated papers. No invented authors.
4. Keep your AI conversations. If you're ever questioned about how you used AI, having the actual conversation history is your best defense. It shows you were asking questions and researching, not generating submissions. LazySusan saves all your conversations automatically across every AI tool.
5. Run your own detection test. Before submitting, paste your work into the same tools your professor uses. If Turnitin's AI detection flags it, rewrite those sections in your own voice. Better to catch it yourself than have your professor catch it.
6. Add what AI can't. Personal examples. Your own data. Opinions from class discussion. Analysis that connects to other courses you've taken. References to specific ASU experiences. These are things no AI can generate and they're what make your work unmistakably yours.
7. When in doubt, disclose and ask. Seriously. Email your professor: "I used Perplexity to find sources and Claude to help me understand [concept]. Is that within your course AI policy?" Professors respect this. It shows integrity, which is the whole point.
8. Use AI to improve your process, not skip it. The students who benefit most from AI are the ones who use it to work better, not less. AI-generated flashcards from your own notes. AI summaries of readings you've already done to reinforce retention. AI as a study partner, not a substitute for studying.
The tools that help vs. the ones that get you in trouble
Not all AI tools carry the same risk at ASU. Here's a practical guide:
Tools that help you learn (low risk):
Tools that get you caught (high risk when misused):
A note for international students
This section matters and doesn't get talked about enough.
AI detection tools like Turnitin have documented higher false positive rates for non-native English speakers. If English is your second (or third or fourth) language, your natural writing patterns may trigger AI detection simply because they're more formal or structured than typical native-speaker student writing.
If you're an international student at ASU:
Keep detailed records of your writing process. Save drafts, notes, outlines — everything that shows your work is yours. If flagged unfairly, this documentation is your defense.
Know your appeal rights. ASU's Academic Integrity Process includes an appeal mechanism. If you believe you've been falsely flagged, you have the right to contest it. The Office of Student Rights and Responsibilities handles appeals.
Use AI for language assistance transparently. There's a difference between using AI to translate your thoughts from your native language to English (generally acceptable with disclosure) and using AI to generate the thoughts themselves (not acceptable). Be clear about which you're doing.
Connect with the International Students and Scholars Center. They have resources specifically for navigating academic integrity as a non-native English speaker.
The bottom line
ASU isn't trying to stop you from using AI. The university literally partnered with OpenAI. They know AI is part of the future. What they're trying to ensure is that you're actually learning — that when you graduate with an ASU degree, you can do the things that degree says you can do.
The students who use AI well will outperform everyone. They'll research faster, understand deeper, iterate quicker, and produce better work. They'll graduate knowing how to work alongside AI, which is the most valuable professional skill of the next decade.
The students who use AI as a shortcut will get caught eventually. And even if they don't, they'll graduate without the skills their degree is supposed to represent. They'll be the ones who can't perform in interviews, struggle in their first job, and wonder why the degree didn't help.
Use AI. Use it a lot. Use it well. Just don't let it replace the part where you actually learn something.
If you're going to use AI at ASU, use it properly. LazySusan gives you 50+ AI models including Perplexity's Academic mode and Lazy AI with built-in citations — the tools that help you research and learn, not just copy-paste. Student plan: $99/year with your .edu email.