Two Carmel High School students propose an idea to help orient us in the world of AI.

Sara Rubin here—or so you think. How can you really tell if it’s me writing this newsletter column or an artificial intelligence chatbot? 

Spoiler alert: It’s really me. But it’s not always easy to tell, especially on social media, say Carmel sisters Emily and Anna Kim, a junior and freshman, respectively, at Carmel High School. They learned how insidious AI can be when their dad got excited about a musician on Spotify and only later, discovered on social media—not on Spotify—that it was entirely AI-generated. “He was listening to this music for months without knowing,” Emily says.

Their proposal is not to eliminate AI, but to create an alert system so consumers of media know what they are getting. Their simple traffic light pitch is green for minimal AI, yellow for AI enhancements like avatars or voiceovers, and red for fully AI-generated content. They have launched a petition online seeking support for the concept and are hoping to connect to student ambassadors at other schools interested in advocating for the same thing. So far they’ve gotten positive feedback from their peers at Seaside High School and Santa Catalina in joining their new group called Students for AI Transparency.

Both sisters emphasize that AI is not necessarily bad, it’s just that its use should be disclosed. Some uses are obvious—“There are videos of cats cooking,” Emily notes. “When it really becomes a problem is when it is trying to imitate a person or realistic-looking photos that spread false stories.”

They report that essays are now often written in class instead of at home, one way classroom teachers can limit the use of AI. But it can be hard to discern when consuming media. “My sister can tell if it was AI-generated or not,” says Anna, a regular user of YouTube and Instagram. “When I am on social media, I kind of believe what’s on there. I can’t really tell the difference.” 

A simple red-yellow-green would help her, and it would help all of us. 

I can easily imagine a world in which their model or something like it takes off, a warning label for AI. It wouldn’t prevent us from watching ridiculous videos of cats cooking, the same way tobacco warnings don’t stop everyone from smoking. But information is power and the more it flows, the better off we are.