Because instructors strive to promote the analytical, critical thinking, and writing skills needed for all student learning, they expect writing assignments to be the student’s original work. Every submission should be an original composition that students create in response to course assignments. However, because FHSU values a culture of academic freedom and innovation, faculty and staff are encouraged to explore and teach the use of AI in their respective fields.  

Instructor Discretion: Syllabus statements and assignment instructions provided by the instructor always take precedence over these guidelines.  

Disclosure of AI Assistance: If AI tools are used in the creation of academic work, students must clearly disclose their use and specify the extent of the AI’s involvement. This disclosure should be included in a dedicated section of the assignment or as specified by the instructor. 

Recommended Uses of AI: 

• Research: AI can be used for gathering information, generating ideas, and enhancing understanding of academic topics. Please see Research with AI Guidance.

• Drafting and Editing: AI tools may assist in drafting and editing, provided that the final work is reviewed and revised by the student to ensure it aligns with their own voice and comprehension. 

Non-recommended Uses of AI: 

• Generation of Substantial Content: AI should not be used to generate substantial portions of written assignments, research papers, or other evaluative materials. 

• Misrepresentation: Students must not present AI-generated content as their own, nor should they use AI to complete assessments or exams designed to evaluate their personal skills and knowledge. 

Academic Honesty: Violations of this policy, such as presenting AI-generated content as original work without disclosure, will be treated as breaches of academic integrity. Consequences may include disciplinary actions as outlined in the institution's broader academic integrity policies. 

Training and Resources: FHSU will provide training and resources on the ethical use of AI tools and the importance of maintaining academic integrity. Students and faculty are encouraged to seek guidance if unsure about how to appropriately use AI tools in their work.

• Instructors should be clear with students about expectations for the course as a whole and for each assignment. Instructors should let students know whether generative AI use is prohibited, allowed for brainstorming only, and/or allowed to generate content for the final product. Instructors should go over how to cite generative AI use and, if it is a research course, how to conduct research ethically using AI (see below). It is recommended to include a statement on AI usage in the syllabus.
  • For FHSU Department Syllabus Statement Templates: ai/department-ai-guidance
  • For general assistance: https://tigerlearn.fhsu.edu/unlocking-the-potential-of-ai-crafting-syllabus-statements-and-assignment-language/  

• Instructors should be transparent with students about their own generative AI usage. If instructors use generative AI to help write the syllabus, draft assignment instructions, create grading rubrics, provide examples, build teaching outlines or slide decks, write assessment questions, or generate other course content, they should include a citation or disclaimer with the content.
• Instructors should be cautious in using generative AI for grading. Instructors should respect students’ privacy by letting them opt out of having their work shared with generative AI platforms. Instructors should let students know if they plan to use generative AI to help provide feedback on early drafts. Instructors should not solely use AI to grade final drafts without also providing personalized feedback. 

Artificial Intelligence (AI) tools offer unprecedented capabilities to enhance and streamline academic research while being a compelling subject of study themselves.  However, the effective and ethical use of AI in research requires understanding best practices to maximize benefits while minimizing risks.  This document outlines a policy for Fort Hays State University (FHSU) administrators, faculty, staff, and students leveraging AI tools in their research or studying the topic. 

The policy outlined here is a set of guiding principles broken into categories of application.  While it does include some specific rules immediately useful based on the current state and availability of AI tools, its real value is as a guiding framework to help make decisions when confronted with the wide range of choices researchers will face.  No policy on this topic can be comprehensive, so it is recommended this document be interpreted in the following ways when administrators, faculty, staff, and students engage in research in the area of AI: 

• Artificial intelligence is useful and popular, but it is still just one of many tools and areas of research; it should not be placed above or beyond the existing values and rules of FHSU. 
   
• All members of the FHSU community should prioritize the dignity, rights, responsibilities, and intrinsic value of people; when faced with a particular decision regarding AI, people should be protected over and above the interesting or popular aspects of this area of study. 
   
• The use of AI does not negate any existing standards of ethical and honest scholarship; the use and study of AI must take place within the frameworks of the rights of human subjects, intellectual property owners, and content creators just as any other area of research would. 

The categories of application below help to organize this policy by the particular applications of AI while designing, executing, and disseminating scholarly and creative works at FHSU.  

Using AI to Conduct Research 

Researchers should stay informed on the evolving aspects of AI relevant to their work.  This is part of the professional development expected of all faculty. 

Researchers should validate AI-generated data, creative works, information, and results against established methods appropriate to their field and type of research.  AI tools should complement, not replace, traditional research methods by integrating AI data or results with existing analytical approaches. 

Researchers should be aware of potential biases in AI algorithms and data. They must take reasonable steps to ensure research designs include steps to identify and mitigate bias whether technical, epistemological, or cultural. 

Researchers should ensure that data used in AI tools respects applicable privacy laws and ethical standards.  This includes implementing data security measures to protect sensitive information from unauthorized access and breaches in accordance with FHSU human subjects and information technology policies. 

Researchers should collaborate with AI specialists and other disciplines to enhance the robustness and credibility of their research. 

Researchers should encourage student participation in learning and applying AI tools to research projects by including them in faculty projects or providing resources and oversight for students’ own work.  

Conducting Research on AI 

Researchers should attempt to use widely adopted open-source libraries for developing and deploying AI models when possible.  This helps faculty maintain technical and ethical standards while minimizing the burden of maintaining isolated AI tools and data. 

Researchers should use publicly available datasets whenever appropriate to reduce the cost of data acquisition, increase accuracy, and ensure contributions to AI research are as widely applicable as possible.   

Researchers should engage the open source and scholarly AI communities as much as possible to access best practices and up-to-date tools while increasing the contributions made by FHSU researchers. 

Researchers should employ AI services and platforms rather than implement their own instances when possible.  This saves time and money, reduces potential legal liability, and also reduces the environmental impact of intensive computing processes that consume large resources such as electrical power.  Local instances are more appropriate when constructing or testing novel AI technologies and approaches.  When possible, researchers should work to develop resource-efficient algorithms that require less computational power. 

Researchers should be especially careful regarding the use and appropriation of real persons, their identities, and their intellectual property.   

Disseminating Research Created Using AI 

Researchers should maintain transparency in their use of AI.  All research activities incorporating the use of AI must be transparently documented including the tools used, methods employed, and data sources. 

Researchers should explicitly acknowledge the use of AI tools in research publications.  This includes details on the tools and methods used, as well as to what degree AI tools inform the work to be published. 

Researchers should share their data, tools, and methods as appropriate to their research area and field of studies.  Data should be shared in compliance with legal and ethical guidelines while ensuring research is reproducible through sharing of code, data, and methods. 

Researchers should present findings in a balanced manner, highlighting limitations and uncertainties to avoid overstating AI capabilities. 

Researchers should ensure their submitted works are in compliance with any publisher or conference organizational policies regarding the use and dissemination of work using AI. 

Supporting AI Research at FHSU 

Administrators should support appropriate professional development of faculty and staff to ensure they are able to stay informed about the aspects of AI relevant to their jobs and execution of research. 

Administrators should work with faculty and staff to integrate AI best practices into other FHSU organizations, policies, and standards such as the Institutional Review Board, tenure/promotion criteria, university strategic plan, and Student Code of Conduct. 

Administrators should organize institutional resources in support of research using, and focused on, AI through both adapting existing allocations (technology support, professional development funds, etc.) and dedicated support specifically for work in the area of AI (e.g., computing power, software licenses, electrical power, and scalable cloud resources). 

Administrators should support faculty and staff participation in collaborative networks both inside and outside of the FHSU community so researchers may share knowledge, resources, and best practices with other researchers using AI.  To maximize their value, the administrators should support a range of connections including, but not limited to, industry partnerships, university and governmental collaborations, and interdisciplinary programs designed to connect researchers across ideological and methodological lines. 

Ethical concerns related to AI-related or AI-assisted research with human subjects and personal identifiable information remain the purview of the Institutional Review Board. 

Administrators should expand current support for student research opportunities to include specific mechanisms for student participation in supervised AI research projects. 

Administrators should work with faculty and staff to create an additional policy regarding the use and appropriation of persons and/or their identities (and intellectual property if relevant).  A particularly popular and potentially problematic use of AI is the prediction or recreation of a person’s image, voice, actions, and language across a range of media.  While these technical abilities are crucial to FHSU’s ability to conduct important research in AI, there are also ethical and legal concerns to address at the organizational level. 

Administrators should work with faculty and staff, in the spirit of shared governance, to create explicit standards and responses for non-compliance with this policy. 

Generative AI’s Environmental Impact

(A fact sheet prepared by Cheryl Hofstetter Duffy and Claire Nickerson, members of the Fort Hays State University AI Task Force–February 2026)

Generative AI and Energy Usage

Energy consumption related to artificial intelligence is staggering. You consume 5 times more energy when you query ChatGPT instead of doing a simple web search, for example. Such a high level of energy consumption increases carbon dioxide emissions, puts pressure on the electric grid, and fuels climate change, with its disproportionate negative impacts on communities of color and low-income families.

Data Centers

A single data center can cover hundreds of acres, and the pace of data center construction worldwide is exploding, in large part due to increased AI demand. Aside from land-use concerns, these data centers are expected to consume 1,050 terawatt-hours of electricity in 2026. To put that fact into perspective, consider that data centers will move into fifth place on the global list of energy consumers worldwide–between Japan and Russia. Moreover, this rapid pace of data center construction means that newer, cleaner energy technologies can’t keep up, so data centers rely on fossil fuels to fill that gap.

Training New Models

Generative AI requires greater power density than typical computing. New AI models must be trained, and a generative AI training cluster can consume seven to eight times more energy than typical computing. The training process alone for OpenAI’s GPT-3 is estimated to have consumed 1,287 megawatt hours of electricity–equal to the annual power use of 120 U.S.homes–and produced 552 tons of carbon dioxide. And that’s just one generative AI model! Because of a model’s short shelf life under the pressure of ever-increasing demand and competition, new models are constantly being trained and are constantly consuming alarming amounts of energy.

Generative AI and Water Usage

You may have heard that generative AI uses a lot of water–one data center uses as much as 4,200 average Americans each day–but why? There are three reasons: cooling, electricity generation, and supply chains. 

Water Cooling

Your computer heats up when you use it, so it cools itself using a fan (put your hand next to it and see if you can feel the warm air!). However, for large groups of servers, which generate a lot more heat, using just fans is inefficient. So, data centers also use water cooling–either flowing in pipes past overheated server components or as mist to cool outside air used for air-based cooling systems (evaporative HVAC). Both these processes use fresh water to avoid bacterial growth, and water must be continually replaced because it evaporates or collects minerals that cause clogged pipes.

Water Power

In the same way that the servers at data centers get hot and use water for cooling, power plants ALSO get hot and use water for cooling. And, as you read above, data centers use a lot of electricity. So, data centers have a double impact on water usage: once when electricity is generated at the power plant, and again when it is used to power servers at the data center. 

Supply Chains

Building servers requires computer components, which first have to be manufactured. Unsurprisingly, manufacturing generates heat and uses water for cooling. However, it also uses water to remove dust and other contaminants, including hazardous waste. Because server components are sensitive, the water used must be clean, and because the wastewater recycling rate at large manufacturing plants is low, they need new water frequently. 

How to Use Generative AI Mindfully

Tips for All Users

1. Consider whether you need to use AI at all. Is the task something you could easily do with a web search or some help from another human, or would it be very difficult and time-consuming to do by hand? Is it something you actually need, or would it be just for fun?
2. Avoid using AI unintentionally. If you don’t need an AI overview for your search results, put “-ai” at the end of your Google search, or switch to a search engine that makes it easy to turn off AI, such as DuckDuckGo. Look for and disable AI assistants in your software. Check the data and privacy settings for services you use and opt out of letting your data be used for training AI.
3. Choose the right AI tool. Smaller language models are more energy efficient. If you aren’t feeding in a lot of data and don’t need complex computations, try a mini model like ChatGPT nano from OpenAI, Claude Haiku from Anthropic, or Gemini Flash from Google. (Look for a dropdown menu near the top of the page or the text entry box to change models.)
4. Be clear and concise. Longer prompts and bigger words use more processing power, and so does rephrasing your request because you weren’t clear the first time. Before you hit the enter key, review your prompt to make sure you have provided all the necessary information, written clear instructions, and defined your desired output–in as few words as possible. This process is called prompt engineering. 
5. Create reusable resources. This is especially important if you are generating images, which requires more energy than generating text, or video, which requires much more energy. Has someone else already created an image or video you can use? If you need to generate something new, can you post it somewhere others can find and re-use it?

Tips for Teachers

1. Share the above “Tips for All Users” with your students. Their choices can reduce AI’s environmental impact.
2. Educate yourself on AI’s environmental impact. The information on this fact sheet is a good starting point. Delve deeper by checking out some of the additional resources linked below, making yourself a valuable resource for your students.
3. Include environmental impact as a topic when discussing AI with your students. Digital literacy is vital in all disciplines and should include environmental awareness. Specifically, include environmental impact when discussing ethical AI use. In addition to AI’s potentially negative impact on learning outcomes, creativity, and appropriateness/ accuracy, explore its potentially unethical environmental impact with your students. 
4. Teach students the value of using a search engine or (gasp!) actual humans as resources. Consultants in the Writing Center, peer tutors in various disciplines, and, of course, professors can provide support and social connections/networking. Include those non-AI options as recommendations within your verbal and written instructions for assignments or study guides.
5. Be intentional about your own use of AI. Limit your reliance on AI for chatbots, rubrics, automatic grading, image generation, etc. Use AI to complement your traditional teaching methods, not replace them.

Resources 

Bradshaw, P. (2025, June 19). How to reduce the environmental impact of using AI. 

Online Journalism Blog. https://onlinejournalismblog.com/2025/06/19/how-to-reduce-the-environmental-impact-of-using-ai/

Gorey, J. (2025, October 17). Data drain: The land and water impacts of the AI boom. Lincoln 

Institute of Land Policy. https://www.lincolninst.edu/publications/land-lines-magazine/articles/land-water-impacts-data-centers/

International Energy Agency. (2025). Energy and AI. https://www.iea.org/reports/energy-and-ai

Keller, J.B., Donoghoe, M., & Perry, A.M. (2024, January 29). The US must balance climate justice challenges in the era of artificial intelligence. Brookings. https://www.brookings.edu/articles/the-us-must-balance-climate-justice-challenges-in-the-era-of-artificial-intelligence/

NEA (2025, June 20). Environmental impact of AI. 

https://www.nea.org/professional-excellence/student-engagement/tools-tips/environmental-impact-ai

Perron, B. (2025, September 12). How to Reduce Your AI Carbon Footprint. Medium. https://b-r-i-a-n.medium.com/how-to-reduce-your-ai-carbon-footprint-569c810299aa

Privette, A. & Center for Secure Water. (2024, October 7). AI’s Challenging Waters. University of Illinois Urbana-Champaign Grainger College of Engineering Civil & Environmental Engineering. https://cee.illinois.edu/news/AIs-Challenging-Waters

Tips from scientists about conscious use of AI. (2025, September 30). NCAS. 

https://ncas.ac.uk/conscious-use-of-ai-practical-tips-and-thoughts-from-environmental-researchers/

Zewe, Adam (2025, January 17). Explained: Generative AI’s environmental impact. MIT News. 

Massachusetts Institute of Technology

https://news.mit.edu/2025/explained-generative-ai-environmental-impact-0117