Schools must ensure that students are prepared to confront the ethical dilemmas that Artificial Intelligence (AI), like all new technologies, will inevitably present. Humans have created AI and are learning from it. Consequently, AI will have flaws that require skillful interaction to uphold the basic tenets of right and wrong. Ethics are not merely a reflection of transient feelings, a prescription from religion, or an acquiescence to whatever society deems correct in a moment; instead, ethics embody the grounded reasoning of rightness, fairness, and virtue as applied by humans (Velasquez et al., 2010).
Ethics are neither inherited through birth, citizenship, or culture nor understood equally across languages or ethnicities. Therefore, ethics represent the shared concepts that define the boundaries of right and wrong, regardless of differences.
Ethical AI in an Era of Rapid Transformation
Right and wrong are package deals that drive rational decision-making on dilemmas to determine whether or not an action may harm others (Schwartz, 2016). We do not process the world through only rights or wrongs; the human lens always balances new information through already experienced value systems.
Technological advances force users to evaluate systems of human behavior when the companies that create technology abandon their principles in the race for industry dominance (Prentice, 2025). Just as the light bulb positively extended reading hours to improve learning and negatively increased industrial work shifts to expand corporate profits, so will AI push and pull humans to balance its polarized implications. A new frontier of ethical AI is born.
Understand and Promote Appropriate Usage through Effective Prompt Writing
Teachers must model behavior for using AI to minimize unwanted material. We cannot blame AI if its influence is inappropriate, lacking, or unjust. Instead, we must understand that AI crafts responses based on the interaction of its large language model (LLM) training and the inquiries we provide. AI does not only learn from its original training models; it also learns from us and about appropriate use.
LMs depend on expanding their information base through language interaction patterns. Consequently, students who engage in dynamic, thoughtful, and purposeful inquiries with AI may have stronger responses and enjoy more positive interactions. However, inexperienced students’ impulsive, clumsy, or inappropriate inquiries can lead to cascading failures in future interactions. According to Cornell University (2025), “The data used to train the generative AI tool will mimic the data it receives,” potentially resulting in “… inaccurate, misleading, and unethical information,” which, in turn, creates a potentially toxic dynamic that requires instructor modeling for appropriate usage.
Modeling healthy prompt writing for students can ensure a tailored approach to appropriate AI interaction. Just as answering a phone requires linguistic etiquette, engaging properly with AI necessitates decorum. Prompt writing strategies teach students that their inputs affect their outputs (garbage in—garbage out). Four key elements are essential to ensure students receive the intended outputs from AI, promoting learning and preventing the AI from inadvertently learning undesired patterns of inquiry from the students.
Prompts should include:
[1] the context of use (What do you want to know?)
[2] the output restrictions (How much do you want to know?)
[3] the audience receiving the information (Who is interested in the results?)
[4] a defined or refined tone of the discussion (What needs further clarification?)
When engaging in dialogue with an AI, the defining and refining stage is iterative; the AI mimics and adapts to a traditional human conversational experience (Bansal, 2024). It may take several specifications to guide a final response, such as requesting the AI to shorten the reply or to include or exclude specific events or individuals. Prompt writing should follow an ethical decision-making process by avoiding private information, emotionally charged conjecture, or reputation-damaging judgments, as this information may become part of the AI’s overall identity intertwined with a user’s personality. Given that LLMs and AI trainers already outperform humans in identifying personality traits through psychometric analysis of social media users, creating ethical concerns regarding privacy invasion and the development of individual autonomy, we should anticipate that casual AI chats will soon be flooded with personality identifiers. (Peters & Matz, 2024; Schoenegger et al., 2025).
Since student users are still developing their values and social identities, both online and in-person, proficient prompt writing skills may further shield them from unwanted personality association, merging in AI ecosystems.
Return to Pedagogy Basics
Theoretical contributions lay the foundation for pedagogy, establishing the basis for acceptable teaching methods and norms. Until we develop a universal code of ethical teaching practice or new theories that address the complexities of education alongside AI, we are best served by adhering to theories that have already stood the test of time. However, this requires a commitment to professional development grounded in theoretical principles and research-based practices.
Video conferences, prevalent during the COVID era of education, highlighted national school practices unsuitable for learning environments and lacked parental support. Some public schools placed more emphasis on social issues rather than instructional improvement, resulting in decreased enrollment. Of equal concern, private schools reported the highest number of teachers without a bachelor’s degree and faced challenges in retaining certified teachers who meet licensure requirements (NCES, 2021). Consequently, a lack of adherence to and understanding of effective educational pedagogies has pervaded learning environments, raising the question of integrating ethical AI into an already fractured landscape. Regular professional development may offer a solution, as it typically includes curriculum, instruction, or assessment components that align with pedagogy.
When schools facilitate knowledge of theoretical foundations, they guide educators in creating lessons that build learning frameworks, enabling students to progress from one level to another, enhance cognitive connections to environmental stimuli based on rewards or consequences, and improve educational interactions through modeling. The historical tenets of Vygotsky’s scaffolding, Skinner’s conditioning, and Bandura’s modeling may pave the way for ethical AI education.
Cognitive approaches, for instance, support lessons about the stages at which students should receive material based on age and comprehension abilities. While it may seem obvious that different ages bring varied expectations, as every parent understands, trained educators utilize this developmental information to assess whether a learning component aligns with established expectations for a student population. Thoughtfully crafted lesson plans consider diverse stages to ensure understanding is achieved and behavioral issues do not arise from frustration. If frustrations occur, particularly at the elementary level, a behaviorist approach based on conditioning responses to rewards has proven beneficial. These supports often include prize boxes, behavior charts, or fun activities to help complete challenging projects.
As a typical instructional day includes more theoretically supported and research-based practices, the instructional environment becomes intentionally prepared to promote ethical AI awareness. Following this, schools can implement meaningful curriculum changes incorporating ethics education across various subjects and grade levels and as a distinct required course for high school students. Ideally, educators’ deliberate efforts will bridge the knowledge gap on ethical AI practices until new theories or clear ethical guidelines are established.
Require Ethics Education to Promote Resilient Youth Entering an AI Workforce
While learning to use AI is important, educators must be mindful of what information within the AI landscape is suitable for formal education at each age level. As ethical AI is integrated into the school curriculum, special care should be taken to ensure that ethical AI education remains distinct from the general use of AI. General ethical principles, such as privacy, fairness, and trust, should be taught alongside specific ethical AI topics like human–AI collaboration, the distinctions between plagiarism and cheating, and school-based responsibilities regarding AI usage.
Human–AI collaboration encompasses more than merely using AI as a tool; ethical AI education emphasizes teaching decision–making to utilize AI based on varying situations (Adams et al., 2023). Human–AI collaboration will become a necessary component of the workforce, even if individuals choose not to engage with AI in their personal lives. With industry investors projecting fourfold returns on AI investments, all aspects, including employment partnerships, are anticipated to be fully integrated (Avanade, 2024). In college curricula, human–AI collaboration through prompt engineering techniques encourages the development of critical thinking and creativity skills, especially when educators design subject–specific activities such as project redesigns, experiments, business partnerships, or template creation (Lee & Palmer, 2025). This similar skill set development could enhance K–12 classrooms in demonstrating the collaborative process of human–AI dialogue.
Responsibility in AI is often linked to accountability and safety, encompassing both users and creators of AI (Smith, 2024; Jobin et al., 2019). However, consumers have limited control over how AI creators train these systems, leading to controversy regarding who or what should bear responsibility for AI outcomes (Jobin et al., 2019).
Therefore, students must understand that responsibility is the ethical foundation that ensures all other ethics are maintained, focusing entirely on the individuals who interact with AI, whether they are users or creators. Responsibility is an ethical trait currently attributed solely to humans, and as our exponential connections evolve through AI interactions, users must act responsibly to safeguard privacy and well–being (Dignum, 2021).
Since ethical AI education hinges on decision–making, proper ethics instruction must address the distinctions between plagiarism and cheating. Often used interchangeably, these two traits differ in characteristics and intent. Plagiarism is an academic action that lacks proper credit for authorship; it may result from poor etiquette, referencing inexperience, or an intent to cheat.
When plagiarism is considered intentionally dishonest, it is elevated to cheating. Cheating is the intent to inflict harm through dishonest behavior that is broadly deemed unacceptable. Cheating reflects a character flaw, described as “…the exploitation of prohibited materials… to gain an unfair advantage over other students from the teachers” (Sozon et al., 2024), stemming from poor ethical decision–making. A student can be taught to enhance writing or speaking activities to ensure proper attribution. Instruction on cheating, however, is more complex due to its ethical implications.
In self–report surveys, compared to private school students, more than twice as many public high school students and three times as many charter school students reported behaviors associated with cheating through unauthorized AI or devices (Lee et al., 2024).
Given the stark differences in cheating, it is evident that most schools have not fully explored the idea of cheating as an ethical issue. Many private schools have a religious affiliation (Broughman et al., 2021), and their character associations to such may regularly address cheating, leading to fewer instances.
Adolescence is a difficult time when students naturally struggle with many ethical dilemmas; cheating distinctions are no different. Regardless of the school environment where a student attends, if teachers circle back to pedagogically appropriate lessons, including varied assignment types and formative assessment strategies, cheating opportunities may diminish.
Project–based learning, team activities, tiered assignments, flexible grouping, and real–world application are helpful instructional strategies that promote ownership of learning and move the teacher from a disciplinarian to a facilitator role. Students are not relieved of their ethical obligations if they dislike an assignment or if a teacher has not tried a new format.
Final Thoughts
Districts, teachers, and students must collaborate to navigate the rapidly changing learning environment. Ethical AI education is the crucial boundary that imparts essential critical thinking and decision-making skills and must be addressed immediately. Schools can enhance ethics education by evaluating AI-enabled support regarding levels of acceptability. Districts should strengthen professional development to return to effective pedagogy, and teachers must demonstrate appropriate behaviors and elucidate challenging ethical concepts, which will create an environment for students to learn collaboratively with AI. The world has transformed with AI. Will we remain the same?
About the author
