NYU is a big, decentralized university, and its AI rules reflect that. Most of them come from your school, your department or your instructor. A few things are university-wide, though, and they're more student-friendly than you might expect: every student gets Gemini and NotebookLM, and NYU doesn't use AI detectors as proof of anything.
Here's what NYU's official FAQ says, and which models fit the work NYU students do at Tisch, Stern, Courant and beyond.
What NYU actually says
NYU's FAQ on teaching with generative AI is written for faculty, but it's the clearest statement of the university's position. We checked it in October 2026.
Who sets the rules. Most AI rules are set by schools, departments or individual instructors. Students must follow NYU's academic integrity policy and each course's requirements. That policy forbids presenting work "created by another, substantially or in whole, as one's own." The FAQ admits there's no obvious line between acceptable and unacceptable AI use, which is why it asks instructors to state their expectations clearly.
Detectors. NYU doesn't license AI detectors. The FAQ says they're prone to false positives, especially for students who learned English as an additional language, and that their results "should not be treated as proof." NYU turned off Turnitin's AI-writing detection in 2023.
Tools for every student. All NYU students, faculty and staff have access to Google Gemini and NotebookLM through their NYU accounts. The FAQ says Gemini used with an NYU account doesn't train Google's models on your data.
Privacy. Free, non-NYU tools don't give you NYU's data protections. Don't put information covered by FERPA, personally identifiable information or non-public academic data into public AI tools.
What that means in practice
- Your school's rules matter. Tisch, Stern, Steinhardt, Tandon and CAS can all approach AI differently, and so can individual courses.
- A detector score isn't the whole story. If you're ever questioned, your drafts and process matter more than a percentage.
- Use your NYU account for course material. Gemini and NotebookLM through NYU come with protections that personal free accounts don't.
- Keep your process visible. Version history in Google Docs and saved chats are the best evidence that the work is yours.
Which models suit NYU coursework
Tisch: film, TV and dramatic writing
Tisch students write, shoot and edit constantly. Where your instructor allows AI and you disclose it, it helps most in pre-production.
- Script notes. Claude is a careful reader for structure and character notes on your own pages.
- Look books and storyboards. Image models like Nano Banana and FLUX help you show a cinematographer what you mean. Our Nano Banana vs FLUX for design students post compares them.
- Previs. Video models like Veo and Runway can block out a shot before you have a crew.
Try: "Here's a scene from my pilot. Don't rewrite it. Tell me what each character wants in the scene and where that want isn't clear on the page."
Stern: business and finance
For cases and group projects, use AI to test your reasoning, not to write the deck.
Try: "Here's my valuation for this company and my three key assumptions. Which assumption is the analysis most sensitive to, and what would change my recommendation?"
Check every number against the case or the filings. Models get figures wrong.
Courant and CS
For math and CS courses, use AI to understand concepts and debug your own work, where the course allows it. Our best AI for coding classes post goes into detail on which models explain code best.
Try: "Explain why this proof by induction fails at the inductive step. Don't give me the fixed proof."
Research and writing
NotebookLM, which NYU provides, is good for working through your own readings. For finding new sources, a search model that links real pages, like Perplexity, is a good start. Open every source before you cite it.
How to use AI at NYU without breaking the rules
- Check your school's and course's policy. Start with the syllabus, then your school's academic integrity page.
- Ask when the policy is unclear. NYU's own FAQ admits the line isn't obvious, so ask.
- Use Gemini and NotebookLM through NYU for course material.
- Disclose permitted use. Name the tool and what you used it for.
- Keep drafts and version history. It's your best defense against a false accusation.
- Keep FERPA-covered and personal data out of public tools.
Where LazySusan fits
For course material, start with the Gemini and NotebookLM access NYU gives you. LazySusan helps when you need models NYU doesn't provide, or want to compare several at once: a short film outside class, a side business, personal projects.
It brings ChatGPT, Claude, Gemini, Grok, Perplexity and DeepSeek into one workspace with one bill, along with Veo, Runway, ElevenLabs and image models for creative work. Multi-Chat sends one prompt to several models so you can compare their answers side by side.
The Student plan is $99 a year with a .edu email, with 5,000 tokens a month that roll over. Prompt School teaches prompting basics, and you can sign up here.
LazySusan isn't affiliated with or endorsed by New York University. Your school's and instructor's policies come first.
The short version
At NYU, your school and instructor decide how AI can be used, and the integrity policy forbids passing off someone else's work as yours. NYU gives you Gemini and NotebookLM, doesn't treat detector scores as proof and asks you to keep private data out of public tools. Read each policy, disclose what you use and keep your drafts.