One of the main concerns about generative AI is “cheating,” or students getting credit for work they didn’t do. This is actually a problem that collaborative learning has been grappling with for decades. In fact, if you think of generative AI as a collaborator in a group project, there’s actually quite a lot of existing practice and literature we can tap into for guidance about using generative AI effectively in the service of learning - both in how students learn and how instructors assess. ...

I’ve found this to be a useful analogy for understanding what it means for generative AI to be “open”: Operating systems can be proprietary or open source. Device drivers and other kernel modules that change the behavior of the operating system can be proprietary or open source. Software programs that run on an operating system can be proprietary or open source. Foundation models are like operating systems. Fine-tuning a foundation model is like writing a device driver or other kernel module. ...

If you’re not listening to the Latent Space podcast, you’re missing some of the best thinking on generative AI happening right now. The show notes for a recent episode begin, Stop me if you’ve heard this before: “GPT3 was trained on the entire Internet”. Blatantly, demonstrably untrue: the GPT3 dataset is a little over 600GB, primarily Wikipedia, Books corpuses, WebText and 2016-2019 CommonCrawl. The Macbook Air I am typing this on has more free disk space than that. In contrast, the “entire internet” is estimated to be 64 zetabytes, or 64 trillion GB. So it’s more accurate to say that GPT3 is trained on 0.0000000001% of the Internet. ...

Last week I asked a “What if?” question about the way generative AI might change the ways that learners interact with instructional materials like textbooks. This week I’d like to ask another. I’m at the GRAILE workshop on AI and higher education in Denver today, and sitting here in this space I’m hearing things through a slightly different filter. For example, when someone mentioned teaching assistants earlier this morning, it made me think - what if generative AI could provide every instructor with a genuine teaching assistant - a teaching assistant that actually assisted instructors with their teaching? ...

There was some interest in my post yesterday about what I called “generative textbooks,” but based on people’s comments I don’t know that I explained the idea very clearly. The idea of a “generative textbook” is that, instead of containing instructional content, it provides the learner with a series of prompts they can use to elicit information from a large language model like ChatGPT. Here’s a brief example to clarify what I mean. ...

I’ve limited myself to one hour of writing for this post, so it’s more a collection of thoughts than a coherent narrative. I expect I’ll have lots more to say on this topic in the future. For now, I just want to get the beginnings of this idea out into the world, together with some initial implications. Since ChatGPT’s meteoric rise to popularity, I’ve constantly been amazed by the creative ways people have imagined to make use of generative AI. For months now it feels like each week I read about some new way generative AI is being used that just completely blows my mind. For example, my son who is studying cybersecurity recently told me prompted ChatGPT along these lines: ‘You are the hiring manager for a cybersecurity position at a large company. Conduct a mock interview with me, asking me ten conceptual questions. After I have answered all the questions, give me feedback on my answers. Then ask me ten technical questions. After I have answered all the questions, give me feedback on my answers again.’ My son tells me this exercise was incredibly helpful and he plans to do it several more times before having a “real” interview. ...

Our work at Lumen is focused on eliminating race and income as predictors of student success in the US postsecondary setting. One thing we’ve learned as we’ve worked to erase this persistent gap in academic performance is that it is far easier to “slide the gap to the right” than it is to close it. In other words, interventions intended to benefit the lowest performing students often benefit all students, so that everyone’s academic performance improves. That’s great from one perspective - everyone learned more! But rather than decreasing the size of the gap, these interventions leave the gap in tact and nudge it up the scale to the right. Interventions that have an accurately targeted effect can be hard to find. ...

If you can remember the web of 30 years ago(!), you can remember a time when all it took to make a website was a little knowledge of HTML and a tilde account on the university VAXcluster (e.g., /~wiley6/). While it’s still possible to make a simple website today with just HTML, making modern websites requires a dizzying array of technical skills, including HTML, CSS, JavaScript frameworks, databases and SQL, cloud devops, and others. While these websites require far more technical expertise to build, they are also far more feature-rich and functional then their ancestors of 30 years ago. (Imagine trying to code each of the millions of pages on Wikipedia.org or Amazon.com completely by hand with notepad!) ...

At some point over the last decade, open educational resources (OER) advocacy in US higher education became zero textbook costs (ZTC) advocacy. The two are intertwined now in a manner that would be difficult to disentangle even if you wanted to try. There are plenty of practical reasons why this might have happened. For example, politicians understand costs much better than they understand learning, which makes policy work and other political advocacy around eliminating textbook costs far easier than advocating for ways that “open” (whatever that word means) might be leveraged to improve student outcomes. But OER / ZTC advocates have had a fundamental problem simmering for many years now, and the recent advent of large language models (LLMs) like GPT-4 will quickly bring that simmer to a boil. ...

The journal Nature, and all other Springer Nature titles, have updated their Guide to Authors with rules for using LLMs like ChatGPT when writing research articles for the publication. To summarize, the rules say: 1. DO NOT list the LLM as an author, and 2. DO describe how you used the LLM in a Methods, Acknowledgments, or other appropriate section. With a journal as prestigious as Nature having established formal guidelines, I expect other journals will adopt similar rules relatively quickly. ...