Digital Discoveries: Lakeview students learn about AI in computer science class

February 20, 2025

Lakeview Elementary School fourth graders netted knowledge about artificial intelligence this winter by helping a computer fish for information.
 
Students in Craig Eveleth’s class hooked many answers about machine learning on their educational fishing poles. They trained a computer how to make a data set of fish by clicking on pictures of ocean-living creatures. The more times they pressed on photos of smiling fish, the more likely the program would recognize those images.
 
Jazmyne and Wendy both said the lesson about machine learning was fun. They discovered that artificial intelligence (AI) programs are only as good as the training data that people put into them. They found out that human opinions affect how computers recognize data patterns, that humans cause AI systems to make mistakes and that it is critical to always double check what an AI program says.
 
“I enjoyed how we had to do the problems with the fish,” Jazmyne said. “It gave me a little more knowledge.”
 
“I enjoyed working with the fish and picking them out,” Wendy said. “We got to help it know what is the correct answer.”
 
Lincoln Public Schools K-5 Computer Science Coordinator Caitlin Provance said the AI for Oceans lesson was designed to help students feel more confident about machine learning. She said the school district wants to ensure fourth graders are prepared for AI as they grow older.
 
“As AI becomes increasingly integrated into our daily lives – and will play an even bigger role in the future of our students – it’s essential that they understand the science behind it,” Provance said.
 
Fourth graders will encounter machine learning in all aspects of their lives, as the technology is present in everything from smart appliances to digital cameras. Eveleth said helping them become familiar with the why, what and how of AI will give them the tools to understand how best to use it.
 
“I think the scope of AI makes an introductory lesson like this very important,” Eveleth said. “I think learning that AI is fallible and can make mistakes is important, because the general perception is that AI is this super existence that does everything and doesn’t make a mistake, and that’s not true. AI very frequently makes a mistake. Learning how those mistakes happen and why they happen is important for them.”
 
Eveleth first explained to students what machine learning is. For example, if someone clicked on videos of cars eight straight times on a social media application, a car video would likely pop up the next time they opened that app. The technology inside the app took the data it received and made a decision based on human-caused training.
 
Students then helped an AI program learn what a fish is. Everyone marked at least 50 pictures to add to the group’s overall data set.
 
Sometimes photos of non-marine items like socks or apple cores would appear on the screen. If students clicked “no” on those items, then the computer realized that socks and apple cores were not fish. However, if they accidentally clicked “yes,” then the AI would mistakenly think those pictures were just like sailfish, yellowfin tuna or Pacific halibut.
 
Eveleth deliberately made two errors in his data submission to illustrate the point to students.
 
“The AI didn’t make the mistake. The person did,” Eveleth told the class. “AI systems aren’t always right because people program them.”
 
The lesson then asked students to include only green fish in the data set. It was easy to click on pictures of fish that were entirely green, but other images showed fish with red bodies and green fins. Should those be counted or not? The class soon realized that personal preferences of each programmer can affect how AI systems operate.
 
“We always talk about it afterwards: why did you choose this, why did you choose that?” Eveleth said. “There are always a lot of different opinions in the room about what fits the category that they were trying to teach the machine. That always sparks good conversation too.”
 
Jowan said she liked learning how the AI training process works. She said the lesson taught her to always get multiple sources if she was using AI for classroom projects.
 
“I enjoyed that our teacher made sure it was fun,” Jowan said. “AI sometimes can help you with stuff, but sometimes it doesn’t.”
 
Provance and Eveleth said those comments showed why computer science is taught in elementary school. The subject helps children build valuable critical thinking, creativity, collaboration and communication talents. That, in turn, leads to greater self-esteem and well-being.
 
“By teaching computer science at an early age, we equip all students with the skills and confidence to contribute to and shape the digital future,” Provance said.
 
“Computer science does a great job of teaching a way of thinking that makes them better prepared for math, for science, for being better able to construct something for an essay,” Eveleth said. “I do really believe that the tools they get in here by doing all sorts of computer science activities, they take back to their classroom to help them in other subjects. The AI part is just another piece of that.”
 
Jowan said she wants to make more digital discoveries about machine learning during future fishing trips in class.
 
“I learned that AI can be smarter than you, but humans have to program it to make sure that it’s right,” Jowan said. “It was really fun today.”
 
Visit home.lps.org/cte/cs/ to learn more about the many computer science lessons that benefit our students every day at LPS.
 
Do you have a story idea? Share it with the LPS Communications Team by filling out this form!

---