Artificial Intelligence (AI) at Washington State University
Table of Contents:
Introduction
Generative artificial intelligence (AI) refers to technologies that can create or transform content such as text, images, audio, video, and code in response to user instructions. Generative AI systems use machine-learning models trained on large collections of data to identify patterns and generate new outputs.
Common examples include Microsoft's Copilot, OpenAI’s ChatGPT, Anthropic’s Claude, and Google’s Gemini.
WSU encourages the ethical, informed, and intentional use of AI in teaching, learning, research, and university operations. AI can support creativity, productivity, and discovery, but users remain responsible for evaluating its outputs and protecting WSU information.
Dos and Don'ts of Generative AI
Do:
Fact-Check Information
Generative AI applications may provide detailed and quick responses, but it is important to further verify every response by fact-checking with other reliable sources and ensure the information you use is accurate and reliable.
Stay Vigilant Against Phishing
Generative AI can make phishing, impersonation, and social-engineering attempts more convincing and easier to produce at scale. Treat unexpected requests for credentials, payments, sensitive information, or urgent action with caution. Report suspected phishing to WSU’s Office of Information Security and Assurance (OISA) at abuse@wsu.edu.
Maintain Human Judgment
AI-generated content can contain errors, bias, misleading conclusions, or inappropriate recommendations. Apply appropriate human review before relying on AI output, especially for decisions that could affect students, employees, research participants, university operations, finances, safety, or other people.
Understand Your Tools
Gaining a better understanding for the capabilities, input requirements, and output variations of the generative AI applications you use is important for achieving desired results within the application. This will also empower you to align the various technologies with your educational and creative goals.
Don't:
Share Sensitive Information
By default, do not enter sensitive, confidential, regulated, or other nonpublic WSU information into generative AI tools. This includes student, employee, research, health, financial, and institutional data. Some AI tools approved for use at WSU may permit specific categories of data under WSU’s agreements and security controls. Always check the WSU guidance for the specific AI tool and account type before using nonpublic or sensitive information. Never provide credentials, passwords, authentication codes, or other secrets to an AI tool.
Submit, Publicize, or Distribute AI Content Without Human Review
Review AI-generated material for accuracy, quality, bias, confidentiality, intellectual-property concerns, and compliance with applicable academic or workplace requirements. You remain responsible for content you submit or publish, even when AI helped create it.
Generate Harmful Content
Generative AI applications can generate harmful, inaccurate, or offensive content. Uphold responsible AI practices by using these systems to create positive and meaningful content while avoiding biased or harmful outputs. Prioritize ethical practices to establish positive interactions and always keep your core ethical principles.
Treat AI as an Authoritative Source
AI-generated responses are not inherently authoritative, even when they include sources or citations. Verify important information against appropriate primary, scholarly, or institutional sources and exercise independent judgment.
Before Using an AI Tool
AI services can differ significantly in how they store, process, share, and use information. Consumer, free, individually purchased, and WSU-provided versions of the same AI product may have different privacy and security protections. When conducting WSU work or using WSU data:
Prefer AI services approved or provided by WSU.
Make sure you are signed into the appropriate WSU account or environment when required.
Review the WSU-specific guidance for the tool before entering nonpublic information.
Do not assume that a consumer or personal account has the same protections as a WSU-provided version of the service.
Follow applicable WSU data-handling, privacy, research, academic, and information-security requirements.
How to Safely Use Generative AI at WSU
For Faculty
Please review the Artificial Intelligence (AI) for WSU Faculty article.
For Students
Please review the Artificial Intelligence (AI) for WSU Students article.
Cite the Work of AI
If you use generative AI to build your work, appropriately cite AI-generated content by including in-text and reference list citations. If your required style has not updated its guidelines around AI citations, be as transparent as possible about how you obtained your information.
Regulated Data and Research
Protect Sensitive and Regulated Data
By default, do not enter sensitive, confidential, regulated, or otherwise restricted WSU data into generative AI tools. This includes information such as student records, health information, financial information, human resources records, human-subject research data, credentials, and other nonpublic institutional information.
Some generative AI tools have been specifically approved by WSU for use with certain types of data because WSU has established appropriate contractual, privacy, security, or technical protections. When using a WSU-approved AI tool, follow the data-use guidance published for that specific tool. This general precaution applies broadly; tool-specific WSU guidance may provide additional or more specific recommendations.
Before providing non-public WSU information to an AI tool:
Confirm that the tool and the account you are using are approved by WSU.
Review the WSU guidance for that tool to determine what data classifications are permitted.
Use only the minimum information necessary for the task.
Follow applicable WSU policies and any additional requirements for regulated or research data.
If you are unsure whether a tool is approved for the data you intend to use, do not enter the data until you have confirmed the applicable requirements.
Research Support
Generative AI can support research activities such as brainstorming research questions, exploring concepts, assisting with code, summarizing appropriate materials, identifying patterns, and supporting literature exploration.
Researchers remain responsible for evaluating AI-generated results. AI systems may produce inaccurate or fabricated references, quotations, data, conclusions, or other scholarly information. Verify important claims and citations against authoritative sources before incorporating them into research.
Research involving human subjects, regulated or restricted information, sponsor requirements, unpublished intellectual property, or other special data handling requirements may require additional safeguards. Before providing research data to a generative AI service, confirm that the specific WSU-approved tool is permitted for the data and research activity involved.
Additional References
AI tools such as ChatGPT are generating a mammoth increase in malicious phishing emails
Guidelines for Use of Generative AI in Marketing and Communications
Guidelines for the use of Artificial Intelligence in Research/Scholarship/Creative Activities
NIST Trustworthy and Responsible Artificial Intelligence Resource Center