{"id":2771,"date":"2024-03-12T16:15:35","date_gmt":"2024-03-13T00:15:35","guid":{"rendered":"https:\/\/enfascination.com\/weblog\/?p=2771"},"modified":"2024-04-28T22:05:40","modified_gmt":"2024-04-29T06:05:40","slug":"ai-in-the-classroom","status":"publish","type":"post","link":"https:\/\/enfascination.com\/weblog\/post\/2771","title":{"rendered":"AI in the classroom"},"content":{"rendered":"\n

At Davis I teach a course<\/a> that introduces programming by starting with motivation before principles. That means it’s a course about doing fun stuff. And I just added an assignment for playing with the OpenAI API. <\/p>\n\n\n\n

It’s an online course with its own security requirements, and so I regularly have the students submitting video in which they flash their IDs. To play with that habit, I ask them to modify code for OpenAI’s Vision API to give me a <\/p>\n\n\n\n

A {style} {animal} holding its ID up to a webcam.<\/code> <\/p>\n\n\n\n

Here are their submissions. I don’t have all the prompts that students filled in with, but some are in the filenames. With some, I do wonder if all of these submissions are real AI-made images as opposed to mere human-made submissions, but I have no way to confirm, so I can only enjoy the irony of suddenly being concerned that my students are manually doing their homework instead of using AI.

If it looks fun, here are the OpenAI demos, #1<\/a> (text) and #2<\/a> (images and audio) for you to play with. <\/p>\n\n\n