Lake Pend Oreille School District trustees reviewed a proposed policy Tuesday that dictates permitted and prohibited uses of generative artificial intelligence. 

District Policy 8705 specifies that students may use AI in certain cases, but that they must cite their use or risk being punished for plagiarism. Instructors would ultimately be allowed to dictate the extent of AI use in their classrooms. 

“We tried to make a policy that outlines the specifics of it and then really gives the teachers in the classroom levels that they can issue with AI policy,” LPOSD technology director Matthew Brass said. 

The language states that students would be allowed to use AI to help them research topics, brainstorm for projects and summarize concepts to aid their understanding. They wouldn’t be allowed to pass AI content off as their own or feed an AI model content that contains personal or sensitive information that should remain private.

“That's something I don't think people think about with AI as much,” Brass said. 

The policy also specifies how teachers are allowed to utilize AI. If the policy is approved, instructors will be able to use AI programs to support lesson plan development and for repetitive administrative tasks like routine emails but won’t be able to use AI as a sole means for assessing student performance. 

Brass described a green, yellow and red light system in which teachers can determine if they want students to have AI access for all, some or none of their assignments. 

For teachers who want to eliminate AI from their classrooms, district staff can block AI websites on student’s computers. But to really ensure students don’t consult AI for a particular activity, Brass recommended temporarily eliminating technology altogether and going back to paper and pencil. 

“If you really want to assess what people have, you really have to remove them from the possibility of utilizing AI,” Brass said. 

Evidence of AI use is often “really hard to find,” he added. “You can take a piece of your writing and paste it in and say, ‘I'd like you to write about this. Please do it in my style,’ and it will.” 

Tuesday’s examination was the first reading of the policy. Trustees will later decide whether to adopt it.

“I’m sure that this is going to be a constant battle,” vice chair Lonnie Williams said during the meeting. “I can say personally that in my industry, there is a lot of AI development going on, as there is in every industry in the world right now.” 

“There are fantastic tools that come out of it, and there’s going to be abuses,” he added. “It is not going away.”