5 Considerations when using GenAI

GenAI is impacting our information spheres, our critical thinking abilities, academic integrity and the environment.

Misinformation and ethical considerations

Beyond unintentional errors, GenAI lowers the technical barrier for creating convincing, large-scale, and inexpensive fake content. This includes persuasive fake news articles, synthetic images, deepfake audio, and automated social media bots.

• Researching a current events topic and finding fake news articles or fake primary-source photos generated to manipulate public opinion.
• Evaluating the credibility of online sources for a biology research paper because AI-generated science blogs mimic the tone and structure of real peer-reviewed studies.
• Receiving a deepfake voice message or phishing email impersonating a family member, utility provider, or financial aid officer requesting urgent payment.

Cognitive offloading and critical thinking

Technology tools can naturally support human thought; like how calculators perform arithmetic, calendars track appointments, and search engines locate facts and sources. Using these tools to reduce mental effort is known as cognitive offloading. Because GenAI can handle complex tasks like writing, coding, and synthesizing information, over-relying on it can hinder the development of core academic skills, leaving you underprepared for advanced coursework and professional roles.

• Relying on GenAI to summarize course readings instead of developing your own close-reading and analytical synthesis skills can lead to difficulty during in-class discussions or closed-book exams.
• Using GenAI to draft all of your essay paragraphs without practicing structural organization, can result in a weak personal writing voice and an inability to construct nuanced arguments independently.
• A study by Gerlich found a significant negative correlation between the frequency of GenAI tool usage and critical thinking abilities. They found that younger participants showed higher dependence on AI tools and lower critical thinking scores compared to older participants

Gerlich, M. (2025). AI Tools in Society: Impacts on Cognitive Offloading and the Future of Critical Thinking. Societies, 15(1), 6. https://doi.org/10.3390/soc15010006

Intellectual property, copyright, and data privacy

Many commercial GenAI models use user prompts and uploaded documents to further train their systems. Inputting sensitive, proprietary, or personal information can lead to privacy violations and unintended data exposure. University-licensed GenAI models do not use user prompts and uploaded documents to further train the systems. That doesn’t mean that you shouldn’t still be cautious about what you upload to or prompt the GenAI.

• Uploading an unpublished research lab dataset or a professor’s copyrighted lecture slides into  GenAI to generate study guides, violating copyright laws or institutional data policies – even if it is a University-licensed tool.
• Uploading personal identification documents, financial statements, or private medical details into a GenAI tool to help draft an email risks that personal data being stored on external servers.

Academic and intellectual integrity

Policies regarding GenAI use vary across colleges, academic departments, individual instructors and courses. What is considered standard assistance in one course (like brainstorming or grammar checking) may be classified as scholastic dishonesty or plagiarism in another. Your instructors should communicate to you what they consider to be appropriate GenAI use for each class. While heavy workloads can make using GenAI for assignments tempting, submitting this work without human oversight can violate academic integrity practices or compromise learning.

• Using GenAI  to edit and rephrase an essay draft for a history class without checking the syllabus, leading to a Student Conduct Code violation because the instructor prohibited all generative tools.
• Using a cover-letter generator for a job application that produces inflated or inaccurate skill claims, resulting in an awkward interview when asked to demonstrate those skills in real time.

Environmental considerations:

The global infrastructure powering GenAI requires significant energy and natural resources. Training and running models consume considerably more electricity than regular web searches, driving up carbon emissions when power grids rely on fossil fuels.

• Generating repetitive, long-form conversational text instead of running a basic web search can use up to ten times more electricity per prompt than a traditional search engine.
• A UN study reported that the environmental footprints are often regional, but have global impacts including water and energy usage, air quality and electronic waste disposal.