A Campus with ChatGPT: The Ethics Behind AI Text Generation in Education
Expert philosopher and ethicist shares his thoughts on the ethics and pedagogy of AI text generation in the educational process, discussing how school systems should handle the integration of ChatGPT in classrooms.
By Rachel Lin
It’s no secret that classroom tools and technologies are constantly evolving. Students used to bring a pen and paper to class. Then came the practice of typing lecture notes on laptops and inputting numbers on a calculator. Now, learners type questions and essay prompts into artificial intelligence programs like ChatGPT to understand a dense topic.
AI text generators are a type of software that utilizes artificial intelligence to produce written content or copy. With impressive benefits like increased efficiency and speed, quality content and accuracy, and low cost, the use of these tools is growing among educational institutions. According to scholars from Michigan State University, “AI in the K-12 classroom became even more prevalent as learning shifted to and, in some cases, remained online due to COVID-19.” With the popularity of ChatGPT, which gained 57 million users in its first month, educators across institutions are worried about the academic integrity and ethics of AI integration in the school system. Debates on whether ChatGPT should be banned in classrooms are one of the biggest topics for school boards across the country. The New York public school department has recently blocked access to the program on school computers and networks.
Adding to the ongoing thoughts, opinions, and arguments of AI text generation integration in classrooms is Dr. Wesley Wildman. Dr. Wildman specializes in developing policies for handling ChatGPT and other AI text generators in the context of university course assignments. He is a professor of philosophy, theology, and ethics + computing & data sciences at BU’s Center of Computing & Data Sciences and the executive director of the Center for Mind and Culture, where he uses computing and data science methods to address pressing social problems. He recently took to Reddit to host a conversation about managing AI text generation in educational settings, the ethics of using such tools, the risks to academic integrity, and more. The top takeaways from this discussion highlight current sentiments around ChatGPT and draw attention to emerging AI technology.
1) Recognize that chatbots and AI text generators have been pushing misinformation for years.
The spread of mis- and disinformation is not new. Dr. Wildman emphasizes that humans are the ones causing such problems via AI text generators. He shares an interesting experiment…
2) Finding the right use for AI technologies like ChatGPT rather than banning them in classrooms is the best option for educators and students.
Some critics argue that ChatGPT is threatening how students write and think in a university classroom setting. To better accommodate the evolving technological landscape, Wildman suggests that educators may need to shift to new teaching methods.
3) AI text generators not only impact academia, but also the public at large.
Dr. Wildman highlights the risks AI can have on students’ learning processes in the classroom, and also how the technology will disrupt other industries.
4) Listening, with moral awareness, will allow us to better navigate the integration of AI into our society.
Identifying populations that will be impacted by AI and determining which groups are the most vulnerable to emerging technologies will help our world adapt ethically and proceed the development with caution.
5) There is an issue of intellectual property and originality of production when it comes to detecting plagiarism of AI-generated content.
Dr. Wildman highlights two different perspectives on how AI-produced text can be considered plagiarism. Generative pre-trained transformers (GPTs) allow you to own your own queries and text; however, the definition might not stand when you are acknowledging its original production and intellectual property.
6) Though AI may seem like it will be learning everything for us, GPTs help people learn at a pace aligned with current, growing knowledge.
Given the speed of AI development, there is fear that what we teach the next generation will soon become obsolete. Dr. Wildman assures that programs like ChatGPT are good at helping people learn right where they are. Beyond just providing knowledge, GPTs can become companions too, and create valuable personal connections.
7) Our current assumptions of schooling need to be more fluid in order to adapt to teaching younger generations.
As AI becomes prevalent in the education system, the ethical obligation of teachers should always be adapting. Dr. Wildman states how educators should stop assuming that pedagogy is static, and rethink the goals and methods of achieving certain teaching outcomes.
8) As plagiarism rules are breaking down rapidly due to AI generation, teachers should set strategic classroom goals.
Detection of AI text generation is fully based on probabilistic evidence as many programs are improving in variations close to human speech. Dr. Wildman’s classroom “recipe” is to make the use of AI generation impossible where original writing is needed, incorporate the GPTs into all other assignments, and teach students how to use the tool wisely.
9) By depending on AI less for writing and more as conversation partners, ChatGPT and other text generation programs can enhance students’ studies.
Rather than being used to cheat, AI can act as a powerful classroom tool by constructing new ways of teaching students how to be less writing dependent. Dr. Wildman says students should start by asking AI questions in intelligent ways to begin to truly understand what the technology can offer.
10) We need to reconsider our methods for teaching students to think–starting with the K-12 education policy.
K-12 students heavily rely on writing in their curriculum. As AI continues to transform this critical learning phase, Dr. Wildman recommends that teachers need to get ahead by setting detailed goals.
For additional commentary by Boston University experts, follow us on Twitter at @BUexperts. Follow Dr. Wildman on Twitter at @WesleyWildman. For research news and updates from BU’s Center for Computing & Data Sciences, School of Theology, and the Center for Mind and Culture follow @BU_CDS, @BUTheology, and @MindandCulture.
