FAQs

General frequently asked questions about generative AI at the University.

• What is generative AI?
  Generative AI refers to a type of artificial intelligence (AI) that can produce new content based on the underlying patterns and structures of data it has been trained on. Generative AI includes large language models (LLMs) that can generate new and eloquent text, and other generative AI applications can create realistic images, sounds, music, and videos based on user prompts. Generative AI models are typically based on machine learning techniques such as deep learning, and they learn patterns in the input data to create outputs that appear creative or realistic.
• What generative AI tools do we have access to at SBU?
  DoIT offers a suite of tools that protect your information from use in other applications. Personally identifiable information (PII) subject to HIPAA may not be entered into these applications. Explore them further on their AI Tools webpage.
• Are there privacy concerns with generative AI?
  Yes. Many generative AI tools collect information and re-use it to create content for other users. You should not enter sensitive, personal, or proprietary information into a generative AI tool without understanding the specific protections provided by that tool. When in doubt, keep your information to youself.
• Is material generated by AI applications always accurate and free from bias?
  No. Generative AI applications can generate useful material but most of them indicate that their output is not error-free. You must check material generated by AI applications for accuracy before using it. In addition, AI models contain biases that result from bias within training data sets and from the composition of training data sets.
• What types of AI research takes place at SBU?
  For information about research being done in AI, visit the AI Innovation Institute and our AI Inventory.
• Does Stony Brook have an AI policy?
  No. The University is reviewing existing policies to ensure they appropriately address the power and implications of generative AI. The AI Coordination and Communication work group is assisting policy owners to conduct this review. 

Frequently asked questions about generative AI for students.

• Can I use generative AI tools in my coursework?
  Using generative AI in academic work is prohibited in some cases, allowable in others, and required in some instances. Read policies on academic integrity for undergraduates and for graduate students, consult course policies, and when in doubt ask your instructor about the use of generative AI for an assignment. Best practices are to disclose and describe how you used generative AI in a project. Representing the work of others, including material generated by AI, as your own work is academic dishonesty. For undergraduates, see also the Generative AI - What to Know video. 
• How are generative AI tools helpful in studying?
  Generative AI tools can be helpful based on the discipline and/or class. For example, you can ask generative AI to "quiz you" on a certain topic. Chat with your instructor on how best to utlize these tools so that they not only benefit your success, but align with the instructor's guidelines around generative AI. 
• What classes can I take that focus on AI?
  A list of courses can be found in our AI Inventory. 

Frequently asked questions about generative AI for faculty.

• How can generative AI be helpful in teaching?

  CELT's Artificial Intelligence in Learning and Teaching has resources as a jumping off point, and If you would like to discuss further, schedule a consultation with CELT. 

• What concerns are there around the use of generative AI?
  There are a number of concerns around the use of generative AI. If you would like to discuss further, schedule a consultation with CELT. 
• How can I know/detect if students are using generative AI (what detection tools are available)?
  When introducing the assignment, clarify what is and is not acceptable. Then, you can set up an assignment in Brightspace using the Turnitin Plagiarism Detector Tool. 
• How can I design my assignments to facilitate students generating their own work?
  CELT's Artificial Intelligence in Learning and Teaching has resources as a jumping off point, and If you would like to discuss further, schedule a consultation with CELT. 
• What do I do if I think someone used generative AI in an assignment?
  If, after speaking with the student, you believe it is necessary to refer the case to the Academic Judiciary, you would follow the same process of any other instance of academic dishonesty. 
• Can I use generative AI when writing my research proposal?
  Funding agencies have specific guidelines about using AI in preparation of a research proposal and other materials submitted to the agency. Consult sponsor website for specifics. See for instance NSF; NIH 

Frequently asked questions about generative AI for staff.

• Can I use generative AI tools to do my job at Stony Brook?
  Staff may use applications on the DoIT AI Tools webpage to help them carry out their responsibilities. These tools are not for use with personal data subject to HIPAA. Employees are responsible to ensure the accuracy and reliability of all content generated by AI as if it is their own work. An inaccuracy generated by generative AI that you include in your work is your mistake, not the fault of the AI application. 
• Can I put sensitive University data into my personal account for an AI application?
  Use of a personal account for an AI application risks improper disclosre of University Data, and University data should not be placed into AI applications not supported by the University. Refer to Stony Brook's Policy on Data and Data Access and Sensitive Information Classification Policy.