AI focus is shifting from cheating to better learning

“Ambiguous and underdeveloped” is how a new analysis characterizes the guidance school leaders have received as they attempt to adopt AI as a learning tool. Such direction, particularly from state agencies, is urgently needed as superintendents and other education leaders have said most educators don’t yet have the skills to use AI to its full potential in their classrooms.

As the focus shifts from cheating and plagiarism to how the rapidly advancing technology can be used to improve learning, just a handful of states have shared policies to help steer AI’s use in schools, says the latest report from the Center on Reinventing Public Education think tank at Arizona State University.

“AI recommendations, policies, and access to training for educators are, by and large, ambiguous and underdeveloped,” writes the report’s author, Bree Dusseault, the Center’s managing director.

‘Talking Out of School’ podcast: ‘The most complicated budget year ever’

California and Oregon were the first to offer their school districts roadmaps for navigating AI. North Carolina, West Virginia and Washington have since followed suit while Virginia Gov. Glen Youngkin has directed state agencies to begin setting the course for “AI integration throughout education.”

North Carolina’s and West Virginia’s AI recommendations are singled out in the report as the most substantial of the bunch. Both states recognize AI’s educational potential and the need to train students and teachers to use it safely. They also suggest curriculum and instructional resources.

North Carolina “takes a particularly innovative stance on plagiarism,” Dusseault writes. The state encourages educators to rethink cheating and assume that students will use AI for all writing assignments. It also suggests that schools adopt a red-yellow-green “AI Acceptable Use Scale” to determine whether AI can or should be used on an assignment, and to what extent.

“It is perhaps shortsighted to automatically consider all use of AI as ‘cheating,’” Dusseault notes.

West Virginia takes more of a zero-tolerance approach to cheating and will require that students get permission from parents to use artificial intelligence tools at school. The state also recommends that teachers focus on AI literacy, primarily in computer science and career and technical education.

North Carolina, on the other hand, envisions infusing artificial intelligence across the curriculum and provides strategies for teaching AI literacy in each grade level. The state is workforce-focused, noting in its policy that “75% of companies will implement generative AI by 2025” and that artificial intelligence and machine learning will be among the fastest-growing job fields.

But who’s taking the lead on AI?

While all five states offer “inspiration and ideas,” Dusseault contends that superintendents and their teams will have to take the lead on setting their AI strategies and adopting their state’s “broad suggestions” to the needs of their own students.

“Districts will still need to undertake the time-consuming task of determining which policies should change,” Dusseault concludes. “They will continue to have significant discretion in shaping their AI approaches, as these guidance documents are what they say they are: guidance. States plan to update them routinely, meaning that districts will have to do the same.”