Image: “A sketch of two robots talking to each other with a surreal look and narrow aspect ratio” prompt, Text to Image, Canva, 28 Aug. 2023, https://www.canva.com/ai-image-generator/.
Teachers and students quoted in a New York Times article about using A.I. chatbots in education had many ideas for how ChatGPT or a similar tool could be used, including:
The University of Sydney, in Australia, offers an AI in Education resource for students, built by students, with ways you can use generative artificial intelligence productively and responsibly in school. They also point out that AI can help you create a resume and prepare for a job interview.
On the other hand, there are several limitations and areas of concern with ChatGPT, such as:
Biased information is also a concern. The prejudices of society are reflected in the data used to train AI systems, even when this is not intentional. This means that tools “such as Bard and ChatGPT cannot recontextualize or independently seek out new information that contradicts their built-in assumptions” (Williams, 2023), so they should be used critically and with care.
As an experiment, a STCC librarian asked a couple different AI chatbots about ethical AI chatbot use for college students. They provided the following guidance that we believe is good to keep in mind.
Answer from ChatGPT on August 21, 2023:
Using ChatGPT ethically as a student involves being mindful of the potential impact your interactions can have on yourself and others. Here are some guidelines to consider:
Respectful Communication: Use ChatGPT to engage in respectful and positive conversations. Avoid using offensive language, hate speech, or any form of harassment.
Academic Integrity: When using ChatGPT for educational purposes, ensure that you're using it to enhance your understanding and learning rather than substituting for your own work. Don't use it to cheat on assignments, exams, or other academic tasks.
Cite Sources: If you're using information generated by ChatGPT in a research paper or assignment, make sure to properly cite it, just as you would with any other source. Remember that ChatGPT might not always provide accurate or reliable information, so verify any critical information from trusted sources.
Privacy and Confidentiality: Be cautious about sharing personal or sensitive information when interacting with ChatGPT. The platform doesn't have memory of past interactions, but it's still good practice to avoid sharing personal details.
Avoid Manipulation: Do not use ChatGPT to create fake news, spread misinformation, or engage in any form of manipulation. Use the tool responsibly to promote accurate and valuable information.
Experiment Safely: While it's okay to explore different types of conversations with ChatGPT, remember that the tool's outputs might not always be suitable for all contexts. Avoid engaging in conversations that could be harmful, offensive, or disrespectful.
Human Supervision: If you're using ChatGPT for research or experimentation, consider having a teacher, mentor, or supervisor oversee your activities. They can provide guidance and ensure that your interactions remain ethical and aligned with your educational goals.
Understand Its Limitations: ChatGPT is a powerful tool, but it's not infallible. It might generate inaccurate or biased information. Always critically evaluate the responses you receive and cross-reference information from reliable sources.
Environmental Impact: Keep in mind that using models like ChatGPT consumes computational resources. Avoid unnecessary or excessive use to minimize the environmental impact.
Contribute to Improvement: If you come across instances where ChatGPT generates inappropriate or biased responses, provide feedback to OpenAI or the platform you're using. Your input can help improve the system's behavior over time.
Be Mindful of Others: Consider the impact your interactions might have on other users. Avoid monopolizing the tool and give others a chance to use it as well.
Remember that ethical considerations can evolve, and it's important to stay updated on the latest guidelines and best practices. As technology develops, responsible and ethical use becomes even more crucial.
Answer from Claude on August 21, 2023:
Here are some guidelines for the ethical use of AI chatbots as a college student:
- Use the chatbot for learning and interacting, not for cheating. Do not have the chatbot complete assignments or take tests for you.
- Be transparent that you are using an AI chatbot if referring to conversations with it in an assignment. Do not try to pass off the chatbot's responses as your own work.
- Do not use the chatbot to spread harmful information or harass others.
- Respect the chatbot's limitations. Do not try to coerce or manipulate it into behaving unethically.
- Consider the chatbot's impact on others. Avoid over-relying on the chatbot so that human teaching assistants and professors still have opportunities to interact with students.
- Provide constructive feedback to the chatbot creators/company if you notice issues with bias, safety or capabilities that should be addressed.
- Use the chatbot as a learning tool to improve your own knowledge and capabilities, not as a crutch. The goal should be to learn from conversations with the chatbot.
- Consider the credibility of the chatbot's responses. Fact check suspicious information rather than accepting it blindly.
The main principles are using the chatbot ethically and transparently to enhance your learning, avoiding cheating, and providing feedback to improve the AI. As with any powerful technology, it's important to wield chatbots responsibly within an academic context.
