What is Responsible AI and What Does That Look Like in the Classroom?

AI has rapidly become one of the most important technologies of our time. It has changed the way we live and work. Despite this fact, many of us do not understand what AI actually is. AI refers to technology that can perform tasks that require human intelligence such as generating content, making predictions, and solving complex problems. Many AI systems rely on machine learning, a process in which computers learn patterns from data. A commonly used form of AI today is Generative AI (GenAI), which is used to create content such as images, text, and code. 

In 1950, Alan Turing created the Turing test in order to determine whether a machine can successfully pass as a human. While the test remains debated, there is no question that AI plays an important role in our lives and in the classroom.

Does AI belong in the classroom? This is the question that all educators are facing. Whether or not you agree with the use of AI, it is here to stay and it is crucial that students understand that AI is not smart nor is it 100% credible.

AI in Education: Reality, Misuse, and Responsibility

Ever since AI became mainstream, many educators have faced serious challenges with AI such as academic dishonesty and the role of technology in learning. There have been two main approaches on how to deal with it. The first approach is to ignore it. Prohibit the use of it, ask students to submit hand written work. The second approach is to incorporate it into the learning process. The first approach is a losing battle. Whether students are allowed to use it in class does not make a difference on how they are using it at home. 

There is an old saying, “if you can’t beat them, join them”.  AI is here to stay, so why not take advantage of the opportunities that AI can bring to the classroom. The key is to use AI responsibly. But what is responsible AI?  My cohort of responsible AI fellows spent several months coming up with a definition that we can all agree with.  Responsible AI is the process of integrating AI in the classroom to enhance—not replace—learning, while maintaining accountability, transparency, and human oversight. 

One of the main concerns about using AI in the classroom is cheating. This fear is quite understandable. Students do not cheat because they are inherently dishonest. They misuse AI when expectations are unclear or when grades are prioritized over growth. The solution to this problem is a change of mindset. In my classroom, I set clear expectations such as:

• AI is allowed for brainstorming, debugging, and conceptual clarification
• AI is not allowed on exams or during initial skill acquisition
• AI must be transparently disclosed whenever it is used

I encourage students to explain how AI helped them, what outputs they accept and/or reject, and what they change on their own. Students should be able to explain their work. If they cannot, this could provide an opportunity to identify learning gaps unless the student has misrepresented AI‑generated work as entirely their own. I also design assessments that emphasize process, reflection, and iteration. AI can generate responses, but it cannot explain personal reasoning, justify design choices, or reflect on growth. When students are required to show their thinking, shortcuts lose their value. Most importantly, I teach students how AI actually works. When students understand that AI is not smart, not neutral, and not always credible, they are far less likely to trust it blindly. Educating students about the pitfalls of AI is the strongest deterrent to cheating.

CARE Model, a Practical Framework for Responsible AI

To support responsible AI use, I created the CARE Framework, a practical model designed specifically for classrooms. CARE was developed through my work as an educator and Responsible AI Fellow and further developed through real classroom implementation with students. The CARE Framework gives students a clear process for thinking critically about AI tools across content.

CARE stands for:

• Check the Data
• Ask About Impact
• Review and Revise
• Evaluate and Explain

CARE emphasizes that AI decisions are ultimately human responsibilities, not technical inevitabilities.

1. Check the Data
   1. Responsible AI begins with data. Educators should encourage the following questions for students to think while they are exploring the model:
      1. What data is included?
      2. What data is missing?
      3. Where did this data come from?
      4. Does it rely on sensitive or demographic features?
   2. Code.org’s AI Lab, for example, allows students to explore this process of training an AI model by having students select a label (what the AI is predicting) and features (the data the AI uses to make that prediction).
2. Ask About Impact
   1. AI decisions affect real people. In this stage, students consider:
      1. Who could benefit from this system?
      2. Who could be harmed?
      3. Could this reinforce stereotypes or unfair assumptions?
      4. Are certain groups overrepresented or ignored?
   2. This step reinforces the idea that AI is never neutral. Fairness is not decided by algorithms. It is shaped by human choices.
3. Review and Revise
   1. Ethical AI requires iteration. In this stage, students are encouraged to:
      1. Remove unnecessary or potentially harmful features
      2. Shift from identity‑based features to behavior‑based ones
      3. Improve or rebalance datasets
      4. Retrain and test models after changes
   2. CARE intentionally challenges the idea that higher accuracy alone equals better AI. Ethical AI is about thoughtful design, not perfect performance.
4. Evaluate and Explain
   1. The final step is accountability and transparency. Students must:
      1. Test models for accuracy and fairness
      2. Compare results before and after revisions
      3. Clearly explain design decisions
      4. Identify limitations, risks, and trade‑offs
   2. This stage mirrors real‑world practices such as model cards and technical documentation. This step prepares students to think like responsible designers rather than passive users.

Why Responsible AI Belongs in Every Classroom

I believe AI belongs in every classroom, regardless of subject area. It is up to educators to set boundaries and define appropriate use. Using CARE or any model of your choice reframes AI as part of the learning process, not a shortcut.

Open conversations, transparency, and hands-on AI experiences reduce misuse and build trust. When students understand how AI works and recognize its limitations, they use it more responsibly.

So, what is responsible AI? Responsible AI is not about control. It is about understanding, ethics, and intentional teaching. AI will continue to reshape education. We are experiencing the growing pains of that transformation. I am optimistic while remaining realistic. AI does not decide what is fair. People do. And that responsibility begins in our classrooms.

About the Author