At this point it would be great if we could just leave AI writer-bots in the dust, much like whatever was at the forefront of last week's news cycle. Although the prospect of taking the tasky part of writing off our shoulders sounds attractive, it's compounding the problems for those of us who are teaching writing. While changes of this nature often lead people to clarify why we are even doing what we're doing, those discussions are usually pretty chaotic, with more talking than listening, more theory than action.
The thing is, for more than twenty years, we have been throwing technology at education. We have entire industries built on it, and career paths, and books and talks and podcasts, and all kinds of material investments in instructional technology. Implicit in all of this is the goal of making learning more accessible, more exciting, more valuable (and maybe more valued), less mysterious, less tedious, and ultimately evocative of the sort of wonder that we attribute to having new eyes and a fresh perspective.
ChatGPT and its compatriots are doing few to none of those things. Whether it's because of rush to make widgets like GPTZero, or the shiny optimism in the infomercial for Copilot, or the scammy gold rush to make apps or sell people on teaching them how to use a relatively simple tool, it's just become more noise. Add to that the fact that, as I've said many times, many of my students had their formative schooling years interrupted and sequestered to screens, and they don't want to stay there. When I hear about all the promise of AI in education, I cringe, because I can conjure up the best days I've had in the classroom. None of them have involved the presence of screens.
"The Age of AI has begun," Bill Gates on his blog, March 21, 2023
Mostly I paid attention to the gloss of AI in education in this blog entry, although some of the productivity tools Gates mentions would also apply. While there's nothing I hadn't already seen suggested, it's good to hear the teacher-student relationship get a nod (at least), and the question of assuring equity too.
"ChatGPT Thinks Americans Are Excited About AI. Most Are Not."Amelia Thomson DeVeaux and Curtis Yee for Five Thirty-Eight, February 24, 2023
Known for bringing data science into the public consciousness, this site includes an article co-authored (sort of) by ChatGPT. Of course the AI made up some dat, which is a no-go for five Thirty-Eight. The author does refer to a Pew survey about Americans and whether they are excited or concerned about the emergence of AI...or both. Where I fall on that question depends on the day.
"ChatGPT's 'jailbreak' tries to make the A.I. break its own rules, or die," Rohan Goswami for CNBC, February 8, 2023
Unfortunately, some people see a new tool and go straight to weaponizing it. While OpenAI has tried to assure the public that they have parameters that will keep it safe for general use, they only can control their chatbots so much... and hackers and criminals even less.
"Generative AI is changing everything. But what’s left when the hype is gone?" Will Douglas Heaven for MIT Technology Review, December 16, 2022
I have this one here because it feels almost quaint. Written several months ago, it doesn't have the crashing outcry that January and February brought to so many sectors. It does give an overview of several AIs and their tricks and their limits, and some of the quotes seem almost naively optimistic. There are also the seeds of agreements-- a partnership between Adobe's Blender and Stable Diffusion, for example.
"This AI clock uses ChatGPT to generate tiny poems that tell the time," James Vincent for The Verge, April 4, 2023
Maybe we'll see more uses like this-- not so intrusive or shady. Just fun. Even if it hallucinates.
Comments