Teachers wanted kids to stop blindly trusting AI. So they had it draw a map of the world. It did not go well.

AP News reported on a Charleston, South Carolina teacher-training session that's become a viral case study in AI literacy.

An instructor prompted an AI tool to "create a map of the world." The result, projected on a large screen in front of a packed high school auditorium, drew gasps and laughter: Mali was spelled "Mail," Egypt was identified as "Sopth," and Libya was replaced with something simply called "Africa." Dozens of other country names were misspelled or pure gibberish.

"If you ever want kids not to over trust these tools, try the map demo," said Amanda Bickerstaff, founder and CEO of AI For Education, who ran the session. Her organization helps schools draft AI policies and trains teachers and students in safe, ethical, and effective use of the technology.

Thirty-seven states have now published official AI guidance that schools can use as a blueprint. South Carolina isn't one of them, so Charleston County's district built its own approach from scratch -- training that reached both teachers and middle and high schoolers, mixing online modules with in-person sessions on how AI tools actually work and how to use them effectively.

The map demo wasn't the only exercise. Reporting on the training also described teachers walking students through a chatbot fabricating an entire fake research study when asked for help with a paper, plus explicit discussions of AI bias and the risks of sharing personal information with these tools.

"Good AI literacy," said Rebecca Winthrop, director of the Center for Universal Education at the Brookings Institution, "includes knowing when not to use it." Her point cuts against a common assumption in ed-tech: that AI literacy mostly means learning to operate the tools well. Winthrop's framing is that real literacy has to include judgment about when a tool is the wrong choice entirely -- not just skill at using it.

After initially trying to ban AI outright in the immediate aftermath of ChatGPT's launch, a growing number of public schools are now encouraging classroom experimentation instead -- specifically so students can see its flaws firsthand, including its tendency to confidently "hallucinate," or fabricate information that sounds entirely plausible.

Source: https://www.wral.com/news/ap/4fb9f-schools-are-starting-to-teach-ai-literacy-for-many-that-means-helping-kids-see-chatbots-flaws/ (AP News, Jocelyn Gecker and Russ Bynum)

Want practical AI guidance for parents and educators every week?
Subscribe: https://www.aibyage.com/?modal=signup&utm_source=beehiiv&utm_medium=newsletter