A Values-Based Approach to Using Gen AI – Faculty Focus

Generative AI arrived in higher education the way raccoons arrive at a campsite: fast, curious, and already halfway into the snacks before anyone agrees on the rules. One minute faculty members were planning lessons, grading papers, and answering emails with heroic levels of caffeine; the next, they were being asked whether ChatGPT belonged in a writing course, a lab, a discussion board, or possibly nowhere near campus Wi-Fi. The question, though, is not simply whether faculty should use Gen AI. The better question is this: how can educators use Gen AI in ways that reflect their academic values instead of flattening them?

That is where a values-based approach becomes more than a nice-sounding phrase. It becomes a practical framework for making decisions when the technology is moving at breakneck speed and campus policy still looks like it was written while someone was frantically refreshing a browser tab. A values-based approach asks faculty to begin with purpose, judgment, and mission before they begin with prompts, platforms, and shiny new features. In plain English, it means professors should not ask, “What can this tool do?” until they have first asked, “What do I want students to learn, what kind of academic community do I want to build, and where does AI help without hijacking the whole enterprise?”

Why a values-based approach matters now

Higher education is not a content factory. It is a human endeavor built on inquiry, critique, trust, mentorship, and the slow, often glorious mess of learning. Gen AI can support that work, but it can also short-circuit it. Used carelessly, it can encourage outsourcing of thinking, create confusion around authorship, blur lines around privacy, and widen gaps between students who know how to use these tools effectively and those who do not. Used wisely, however, it can help faculty brainstorm lesson ideas, generate first-draft rubrics, create practice questions, support accessibility, model critique, and spark deeper discussions about bias, evidence, and judgment.

The central tension is not technology versus tradition. It is convenience versus intention. A values-based approach helps faculty avoid two equally unhelpful extremes: the panic-driven ban that pretends AI will disappear if we squint hard enough, and the techno-cheerleading that treats every automated feature like it was delivered from the heavens on a cloud of productivity metrics. Faculty need a middle pathone that is thoughtful, disciplined, and grounded in the educational mission.

The core values that should guide Gen AI use in teaching and learning

1. Keep human judgment in the driver’s seat

The first value is the most important: human judgment. Faculty expertise is not a decorative extra. It is the heart of teaching. Gen AI can draft, summarize, remix, and suggest, but it cannot replace the context-sensitive judgment of an educator who knows the discipline, understands the students in front of them, and can tell when an answer sounds polished but is academically flimsy. That means AI should function as an assistant, not an autopilot.

In practice, this looks like using AI to create a rough set of discussion prompts and then revising them for rigor, relevance, and tone. It might mean asking AI for three possible assignment structures, then choosing none of them exactly and instead building a better version that fits course outcomes. It also means refusing to hand over final decisions about grading, feedback, or student performance to a system that cannot fully understand nuance, growth, effort, or disciplinary standards. A machine may be quick, but quick is not the same thing as wise. If it were, vending machines would be running faculty senate.

2. Make transparency non-negotiable

If Gen AI is used in a course, students should not have to play detective. Transparency matters because trust matters. Faculty should clearly communicate whether AI is permitted, encouraged, restricted, or prohibited in specific assignments. They should also explain why. Students are far more likely to engage responsibly when expectations are explicit and rooted in learning goals rather than framed as mysterious, shifting rules from the syllabus dungeon.

Transparency also applies to faculty use. If instructors use AI to generate examples, draft materials, or revise prompts, there is value in disclosing that use when it helps students understand how the class is being designed. This does not mean every handout needs a dramatic confession. It means modeling honest academic practice. When educators show students how to acknowledge AI assistance, verify outputs, and make informed choices about attribution, they are teaching more than compliance. They are teaching intellectual responsibility.

3. Put equity and access at the center

A values-based Gen AI strategy must address equity. Not all students have the same access to paid tools, stable internet, strong devices, or prior exposure to AI systems. If faculty assume universal access, they risk rewarding tool familiarity instead of learning. That is not innovation. That is a digital version of “bring your own ladder and good luck.”

Equity means designing assignments that do not penalize students who lack premium tools. It means offering alternatives when AI use is optional, teaching AI literacy explicitly instead of assuming students magically absorbed it from the internet, and recognizing that multilingual learners, disabled students, and first-generation students may experience both benefits and risks differently. Done well, AI can support accessibility and personalized practice. Done poorly, it can deepen existing disparities under the glamorous banner of efficiency.

4. Protect academic integrity without reducing it to cheating talk

Academic integrity is broader than catching misuse. It includes authorship, attribution, honesty, consent, and respect for intellectual labor. A values-based approach does not begin by turning every classroom into a low-budget detective show. It begins by clarifying what counts as acceptable assistance and what undermines the purpose of the work.

For example, using AI to brainstorm research questions for a proposal may be acceptable in one class, while using AI to draft the final analysis may defeat the learning objective entirely. In a coding course, AI-assisted debugging may mirror workplace practice. In a foundational writing course, asking AI to produce the essay may erase the very skill being taught. Faculty should tie AI policies to course outcomes: what students need to practice themselves, what they may do collaboratively, and what kinds of assistance must be disclosed. When expectations are linked to purpose, policies become more coherent and more defensible.

5. Choose critical thinking over convenience

Gen AI is very good at sounding confident, which is wonderful if you enjoy being wrong in complete sentences. That is exactly why it can be pedagogically useful. Instead of treating AI only as a shortcut, faculty can use it as an object of analysis. Students can compare AI-generated answers with scholarly sources, identify hallucinations, test weak reasoning, revise generic prose, and evaluate where the tool succeeds and fails. In other words, the technology can become a lab for critical thinking rather than a vending machine for passable paragraphs.

One strong example is asking students to create a rubric and use it to evaluate AI-generated responses. Another is requiring students to submit both an AI draft and a revised human version, along with a reflection on what changed and why. These activities shift the focus from “Did AI do the work?” to “Can the student exercise judgment, critique, and disciplinary reasoning?” That is a much more educational question.

6. Respect privacy, consent, and data stewardship

Faculty should be cautious about what gets entered into AI tools. Student work, private records, unpublished research, and sensitive institutional information should not be casually dropped into public systems just because the prompt box looks friendly. It is still a data environment, not a diary with excellent autocomplete.

A values-based approach requires educators to understand institutional guidance on privacy, approved tools, and data protection. If a university provides a vetted AI environment, that matters. If it does not, faculty should avoid uploading identifiable student data, confidential materials, or anything that could expose learners or colleagues. Ethical AI use in education is not only about outputs. It is about inputs too.

7. Experiment, but with guardrails

Faculty do not need to become AI evangelists, but neither should they feel pressured to become permanent holdouts. Responsible experimentation is part of good teaching. The key word is responsible. Small pilots, reflective use, and discipline-specific testing are much better than campus-wide chaos disguised as innovation.

A faculty member might test AI-generated quiz questions in one low-stakes module before using them more broadly. Another might try AI as a brainstorming partner for case studies while keeping final content creation firmly human-led. Departments can compare notes, share examples, and develop common language for attribution and acceptable use. A values-based approach welcomes experimentation because it also insists on evaluation. If something saves time but weakens learning, that is not success. That is just faster disappointment.

8. Remember sustainability and restraint

This value gets less attention, but it matters. Not every teaching problem needs an AI solution, and not every convenience is worth the tradeoff. A values-based approach includes restraint: using Gen AI when it meaningfully supports learning, accessibility, or efficiency, and skipping it when it adds little beyond novelty. Sometimes the smartest educational decision is not “How can I automate this?” but “Should I automate this at all?”

That mindset keeps faculty from treating AI like a magical seasoning to sprinkle on every assignment. Some tasks are improved by automation. Others are improved by struggle, iteration, and human conversation. Education still needs room for all three.

What values-based Gen AI looks like in real faculty practice

In a writing-intensive course, a professor might allow students to use AI for brainstorming and outlining but require all drafted prose, source integration, and argument development to be their own. In a nursing course, an instructor might use AI to generate patient scenarios for practice, then have students identify what the scenario gets wrong or oversimplifies. In a history seminar, students might compare an AI-generated summary of an event with primary sources and explain what nuance disappears. In a business class, students could use AI to produce a draft memo, then revise it for audience, ethics, and factual accuracy. The common thread is that AI use is shaped by learning goals, not by novelty.

Faculty can also model reflective use by sharing their own process. They can say, “I used AI to brainstorm possible case examples, but I rejected half of them because they were too generic.” That kind of candor teaches students that effective AI use is not blind acceptance. It is selection, refinement, verification, and accountability.

How institutions can support this approach

Individual faculty members should not have to invent the entire ethical framework alone at 11:47 p.m. while rewriting a syllabus. Institutions need to help. That means offering practical guidance, approved tools, privacy standards, attribution norms, faculty development, and cross-disciplinary spaces where instructors can share what is working. It also means involving educators in policy design instead of dropping rules from a great administrative height.

Strong institutional support does three things well. First, it clarifies principles: purpose, equity, privacy, human oversight, and academic integrity. Second, it gives faculty examples that can be adapted locally instead of imposing one-size-fits-all mandates. Third, it treats AI literacy as an ongoing professional skill, not a one-time webinar everyone forgets by lunch. The institutions most likely to use Gen AI well will be the ones that lead with mission and governance, not just procurement.

Common mistakes to avoid

The biggest mistake is adopting AI because everyone else seems to be doing it. A close second is banning it without teaching students how to understand the technology they will meet outside the classroom anyway. Other common errors include vague syllabus language, overreliance on AI-generated teaching materials, ignoring privacy concerns, and assuming that polished output equals good learning. Spoiler alert: it does not.

Faculty should also avoid replacing authentic feedback with generic AI commentary that says a lot while somehow saying nothing. Students can tell when feedback is specific, and they can also tell when it sounds like it was generated by a machine that has never met a thesis statement it did not find “interesting and multifaceted.”

Conclusion

A values-based approach to using Gen AI gives faculty something better than a trend forecast. It gives them a compass. Instead of asking whether AI is good or bad in the abstract, educators can ask sharper questions: Does this use support the learning goals? Does it preserve human judgment? Does it promote equity? Does it respect privacy? Does it teach students to think more critically, not less? Does it strengthen the academic community rather than hollow it out?

That is the real work. Gen AI will keep evolving. Campus policies will keep changing. New tools will keep appearing with names that sound like startup bingo cards. But values endure. And in higher education, that is the point. Faculty do not need to surrender to the machine, nor do they need to run from it in terror clutching a stack of blue books. They need to lead with purpose, use the technology selectively, and keep the human side of learning gloriously, stubbornly alive.

Experience: What a values-based Gen AI journey actually feels like in practice

In real faculty life, a values-based approach to Gen AI usually does not begin with a triumphant strategic plan. It begins with mild suspicion, a few awkward experiments, and a lot of side-eye. An instructor hears colleagues talking about AI-generated rubrics, faster lesson planning, and tutoring support, then tries a prompt or two late at night after grading. The first result is often a strange mix of useful structure and spectacular nonsense. That moment is important. It reminds faculty that Gen AI is neither a miracle worker nor a guaranteed disaster. It is a tool that becomes educationally meaningful only when it is filtered through judgment.

Many faculty experiences follow the same pattern. First comes curiosity: “Can this help me save time on the repetitive stuff?” Then comes concern: “What exactly is it doing with my inputs, and why does this sample discussion prompt sound like it was written by an overly cheerful robot substitute teacher?” Then comes adjustment. Instructors start using AI for small tasks that do not compromise learning goalsbrainstorming examples, generating practice questions, rephrasing instructions for clarity, or outlining a lesson sequence. They begin to see that the real gain is not perfect output. It is a faster starting point that still requires expert revision.

The most valuable experiences usually happen when faculty bring students into the conversation instead of pretending AI does not exist. A professor might show students an AI-generated paragraph and ask them to diagnose what is bland, misleading, unsupported, or stylistically empty. Suddenly, the classroom energy shifts. Students are no longer just consuming AI. They are interrogating it. They become editors, critics, and thinkers. That is when a values-based approach stops being a policy slogan and becomes a teaching method. The instructor is not asking students to worship the tool or fear it. The instructor is asking them to examine it with discipline-specific judgment.

There is also a relational side to these experiences. Faculty who talk openly about acceptable use, attribution, and learning goals often discover that students respond better to honesty than to vague threats. When students understand why AI is limited in one assignment but welcome in another, they are more likely to see integrity as part of learning rather than just rule-following. Faculty, in turn, feel less pressure to police every keystroke and more freedom to design meaningful work. That does not eliminate misuse, of course. College students will remain college students, and shortcuts will always have fans. But clarity lowers confusion, and confusion is usually where bad decisions love to rent an apartment.

Over time, many instructors describe the same conclusion: a values-based approach makes Gen AI feel manageable. It replaces panic with principles. It allows experimentation without surrender. It helps faculty protect what matters mosthuman judgment, student growth, disciplinary rigor, and trustwhile still making room for thoughtful innovation. The experience is not flashy. It is iterative, imperfect, and occasionally annoying. In other words, it is exactly like good teaching.