I really like the idea of launching an inquiry using See, Think, and Wonder. I especially like this as you could use an image of, say, the Salish Sea. Have students identify the different things they saw, thought, and wondered, and then have them launch independent (or in groups!) inquiries that all pertain to a larger topic area. This way, creativity and wondering can be fostered, while as a teacher you can develop and work towards a culminating goal with the input of your whole student body.
This is an image I had copilot create, which I am basically delighted with. It looks like claymation! I also had it create the image on the top, which doesn’t really resemble Hal from 2001, but it got the message across. To wit: technology doing homework. The image below has no function or raison d’etre other than being very obviously deeply cool if you are fantasy inclined. Who is this dwarf? What is his story? Where did he get such a cooool crown/helmet? Is he worried about the skeletal behemoth lurking in the background? NOPE
I think that, in order to avoid creating inequities, I would like to utilize a spectrum of assignment types and resources. For example, I might begin by having students handwrite the beginning of a project. I know that it will be slow and laborious, but I want to ensure that they are able to get the bulk writing portion done independently. Then, moving onward, we might move into some typed drafts, and eventually culminate in a final draft. Like we did in this class, I think I would create a rubric of allowed AI, and ask for citations. My hope is that, having done a hand-written first draft, I would have a sense of the direction of the project, and could detect if it suddenly shifted in tone or tenor.
UVic, linked above, asks that NO AI be used unless specified by the prof. The challenge here, I think, is that the prof is now tasked with identifying or detecting potential plagiarism. I’m not sure what a solution is, but I imagine that prof’s lives will be made easier by allowing the use of some AI, within defined parameters. I remember during my first degree that, during a midterm, we would write an essay out by hand – probably 800-1200 words, in one sitting. I’m not sure if this sort of thing happens anymore, but it is a good indication of where your comprehension and writing levels are at, without any potential internet aid. If you find that a student can’t put a coherent argument together on paper, and is then submitting some serious work concerning free-will in the context of a multi-verse, maybe your alarm bells are ringing. In one universe they are, fir shir.
The next question is about whether AI might short-circuit learning objectives. I think that this could happen, it totally could. (a study on whether AI might impact cognitive function! More research is needed!) But I think that this is true of a great many tools and inventions: I imagine back when all carpentry was done by hand, and hand tools were utilized, there was probably some concern with the coming of band-saws and table-saws. Almost certainly, some skill and finesse was lost, but then new skills and finesse(s?) would have followed. I think that, for educators, thinking of AI as a tool to be used (and abused, sure!) is a better framework than banning it, or trying to fight it, or whatever. It is most likely here to stay, and understanding it as a tool, with strengths and weaknesses, will probably best set you up for success in the field.
I imagine that AI has some huge potential for language revitalization. Firstly, it can be taught, which should mean that language is sort of digitally *remembered* even if it is not being spoken that much. Then, hopefully, interested learners can engage with AI to attempt to learn, instead of having to potentially move or relocate their lives in order to have opportunities to speak this language. Of course, it wouldn’t substitute for speaking with an actual person, and the amount of nuance and context that you can read from body language and tone. However, as a tool, it seems like a good one to have in the toolkit in the preservation and revitalization of languages.
I seem to be saying tool alot. I am today thinking of AI as a tool to be used.
The exercise we did today asking AI about something we are knowledge about was interesting, as it did take a while to ferret out exactly what it was saying, and what was true. Also, I pointed something out to it and it admitted making a mistake, even though I am not sure it did – I think the language I used might have not been clear enough. But I imagine doing some of these sorts of checks as a class might be an enjoyable way of showing that AI, while a useful TOOL, is very capable of making mistakes. The onus is, and has to be, on the student to check the information and ensure that they are being diligent in their studies and assignments.
I asked Copilot to create the cutest image ever created, and it gave me four nightmare cats in crowns standing in front of a rainbow with flowers. So, that’s a place to leave a blog post.
This thing drinks souls, man, it is *scary* not *cute*
