Helpful Or Harmful? How AI Shaped Education In 2023
Artificial intelligence is rapidly changing the classroom. Students and educators weigh in on the way forward.
When artificial intelligence went mainstream with the public release of ChatGPT in November 2022, it immediately sent ripples through the world of education. Concerns about equity in AI, data privacy, and how to teach ethical AI skills became heated topics at faculty meetings and in classrooms. Two things have become clear: This technology is already transforming education—and people have many questions.
Science Friday hosted a virtual conversation with experts in AI and education in October 2023 to understand the potential AI holds for education and some of the pitfalls to be confronted as it comes to schools across the country. While the path forward may be challenging and occasionally confusing, our guests suggest that by allowing students to lead and guiding them to use AI responsibly, education stands to benefit from innovative tools that will, no doubt, change our world. Here, we recap some of the ideas and insights that came out of that conversation.
A recurring theme around AI in education is that students could use these generative tools to complete their assignments, especially written assignments. Educators are worried that students may not build essential skills if it becomes too easy to misrepresent machine-generated work as their own. In addition to concerns that such use may contribute to cheating, it’s also known that AI can “hallucinate” by returning misinformation that may mislead students. As a result, many schools have banned generative AI completely. But is that realistic?
The answer is no, says Sebastian Rao, a Virginia STEM magnet school senior who has studied the use of AI by students. Instead, he advocates for AI policies that provide “guardrails” for students and educators to make the technology accessible but provide appropriate boundaries. “How do we structure a classroom that uses [AI]?” asks Rao. He says the focus should be to thoughtfully integrate AI technology into classrooms so that students can use it to enhance their education and build new skills.
When schools place strict bans on AI, students (and educators) may be confused and unsure about what they can and can’t use. “We’re having a moment now in education,” notes Andy Forest, founder and executive director of Steamlabs, a non-profit organization focused on innovating around education and equitable access to STEM. “AI tools are just suddenly so capable and competent they are completely disrupting education.” Many students use AI in ways educators and administrators didn’t predict, compounding the problem of regulating their use in the classroom. Meanwhile, the technology is evolving quickly and becoming integrated into an ever-growing range of tools.
Academic tools like Grammarly now use AI, as do search engines like Google and Bing. Khan Academy recently released a free AI-powered tutoring app, Khanmigo, to provide one-on-one tutoring and homework help to students who may not otherwise have access to a live tutor, providing academic equity for low-income learners. “A lot of students don’t really know what to think about a lot of these technologies. Some are confused about, you know, ‘What does AI mean? Is it cheating in certain circumstances?’” explains Rao.
Rather than framing the use of AI as cheating, the U.S. Department of Education is integrating AI into policy because the technology allows educators to support students with a valuable, personalized approach to learning. Knowing how to use AI effectively and ethically will be necessary for students’ future careers. Even the White House is on board. President Biden wrote in a recent executive order that AI will make the world “more prosperous, productive, innovative, and secure.”
AI will almost certainly change how education looks in the classroom, just as computers and the Internet did decades ago. Krystal Chatman, an Instructional Technology Facilitator in the Jackson Public School District in Jackson, Mississippi, believes educators must develop a growth mindset and let students take a leadership role in policy and implementation. “[B]e comfortable taking a backseat to the knowledgeable student,” Chatman advises educators. “You’re used to being the person they seek knowledge from, but teaching them that they too can be experts provides them with a skill that will take them throughout life.”
Students need to be part of the discussion to understand digital citizenship and where the boundaries lie between, for example, using a tool for brainstorming and being academically dishonest. AI can’t and won’t replace human educators. Students will always need adult guides to help them use AI safely. The key to implementation will be developing a nuanced approach where policy creates an environment that protects students’ privacy and ensures academic integrity.
For all its benefits, AI technology does bring challenges that go beyond cheating or ethics. Many parents, educators, and students are concerned about privacy and safety when using AI tools. Forest explains that governmental regulations on data privacy and intellectual property haven’t caught up to the explosion of AI applications becoming available. “[J]ust continuously ask… ‘What are they doing with my data?’ The more you know about AI and how it can use your data, the better because then you can evaluate those policies,” he says.
There are also many concerns about disparities created by AI for students of minoritized racial backgrounds, students with disabilities, and students who are multilingual. Rao explains that students are concerned about AI detectors because the software can incorrectly label written works as AI-generated when they are not, especially for people whose primary language is not English. That may lead to some students being unfairly accused of cheating, which can derail their education or subject them to stigma. However, the AI detectors can be improved and trained on diverse texts to be more representative and inclusive. “Hold these companies accountable,” Rao says.
Chatman agrees and suggests that any accusation of cheating with AI must include humans who can step in and ensure accusations don’t do unnecessary or unjust harm to students. “I know my students better than any platform ever could,” she says. Perhaps most importantly, Chatman stresses that students must advocate for their rights and lead the way. “[W]hen there’s a school board meeting, you can talk directly to the superintendent, with your school principal, your school leadership. Ask them, ‘How are you protecting us?’”
Are you curious to learn more about the role of artificial intelligence in education? Science Friday hosted “YES! You Can Use AI for Your Homework (And Other Stuff Too)” to explore how artificial intelligence is changing education. You can watch the entire discussion on YouTube.
Sandy Roberts is Science Friday’s Education Program Manager, where she creates learning resources and experiences to advance STEM equity in all learning environments. Lately, she’s been playing with origami circuits and trying to perfect a gluten-free sourdough recipe.