I have a seminar meeting coming up for my technology and innovation class, and I’m not sure what readings I should assign. The subject is AI, and I have many thoughts but no conclusions. So today I’m asking for readers’ suggestions. Yes, I’m firing up a good, old fashioned bleg post.
To explain: I teach in Georgetown University’s Learning, Design, and Technology (LDT) program. This class, Technology and Innovation by Design, is required for all LDT students. It takes them through a range of ways of understanding tech, from historical accounts to imaginative writing, economics to critical theory, sociology to feminism, antiracism to area studies. I love teaching it and am always looking for ways to make it better and more exciting. (see latest syllabus at end of this post)
This year I’ve expanded the unit on AI because of the current large language model (LLM) revolution. We’ve done some AI exercises during the semester, and now we’re going to devote hours to the topic.
Midjourney’s best attempt.
For the class session (and maybe two) I have some plans. First, an informal and quick series of go-rounds where I ask students to share what they know about generative AI, what they think about it, then what they’ve done with the stuff. Second, students will do hands-on work with various tools, including chatbots, image creators, and others. That will include discussions and lots of peer learning. Third, I’m going to break habit and lecture for a bit, based on my AI work. That’s to give them a sense of what I see as rapidly emerging things happening, along with a grounding in the technologies.
Besides all of that, and before all of it, I’d like the students to read mind-expanding and interesting material which should challenge them, which is where you come in. Readings should be scholarly, preferably, as one goal of the program is to acculturate students to reading academic research.
Some parameters:
- The students are brilliant. They are mostly international. Age is mostly in the 20s. Gender is balanced.
- They don’t have a lot of coding background, generally, being more grounded in media work.
- So far they have expressed interest in AI, but have not revealed a great deal of experience or enthusiasm for the tech.
- They come from a variety of disciplinary backgrounds, with an emphasis on the social sciences and humanities.
- While the class is focused on technology, the end point is its use in higher education.
- I love historical documents, but want to make sure we include current writing which addresses emerging AI.
I’ve been considering some items with a diversity of approaches and themes, tending to be from the past couple of years:
Brent Anders, The AI Literacy Imperative.
Emily M. Bender, Timnit Gebru, Angelina McMillan-Major, and “Shmargaret Shmitchell,” “On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜”
Ian Bogost, “ChatGPT Is About to Dump More Work on Everyone” and “ChatGPT Is Dumber Than You Think”
James Bridle, “The Stupidity of AI”
Ted Chiang, “ChatGPT is a Blurry JPEG of the Web” and “Will AI Become the New McKinsey?”
Tyler Cowen and Alex Tabarrok, “How to Learn and Teach Economics with Large Language Models, including GPT “
Jason Fagone, “The Jessica Simulation”
Paola Ricaurte, “Artificial Intelligence and the Feminist Decolonial Imagination”
Joana Varon and Paz Peña, “Artificial intelligence and consent: a feminist anti-colonial critique”
Also, using this game for the paperclip problem.
What would you recommend, o readers? Some of these, or others? Thank you for any suggestions.
DALL-E envisions the class.
PS: Here’s the class syllabus so far, for context. I’ve removed some things which don’t seem relevant and anonymized students, as I haven’t asked permission to use their names:
Tuesday, August 29, 2023 – introductions
Readings:
- Plato, excerpt from Phaedrus (the story of Theuth)
- Etymologies (On “tek” just p1)
- The huge technological takeoff (one, two)
Tuesday, September 12, 2023 – Histories of technology, I
- Reading: How We Got To Now part 1: to p. 124
- Introduction to printing press as transformative technology
Tuesday, September 19, 2023 – Histories of technology, II
- Reading: How We Got To Now part 2: to the end.
- On WordPress with Lee Skallerup Bessette
Tuesday, September 26, 2023 – Imagining innovation
Readings:
- Bush, “As We May Think” (https://www.theatlantic.com/magazine/archive/1945/07/as-we-may-think/303881/)
- Schroeder, “Noon in the Antilibrary” (https://www.technologyreview.com/s/611829/noon-in-the-antilibrary/)
- Atul Gawande, “Slow Ideas” https://www.newyorker.com/magazine/2013/07/29/slow-ideas
- Forster, “The Machine Stops” (https://www.cs.ucdavis.edu/~koehl/Teaching/ECS188/PDF_files/Machine_stops.pdf) (https://en.wikisource.org/wiki/The_Machine_Stops)
Student tech presentations: the toilet
Tuesday, October 3, 2023 – How innovations spread, I
- Readings: Everett M. Rogers, Diffusion of Innovations, 5th Edition: 1-52; 72-3; 87-218 (chapter 1; chapter 2 through the Miracle Rice story, the STOP AIDS story, and from “Opinion Leaders” on; chapters 3-5)
- Referenced: Moore, Crossing the Chasm
- On WordPress with Lee Skallerup Bessett
- Georgetown Domains
-
First time Jim Groom talks about Domain of One’s Own
-
Groom, History of UMW Blogs
- Martha Burtis, Neither Locked Out Nor Locked In
Student tech presentation: vaccines
Friday, October 6 – analysis of one innovation due
Tuesday, October 10, 2023 – – How innovations spread, II
- Reading: Everett M. Rogers, Diffusion of Innovations, 5th Edition: chapters 7-11
- Christensen, Raynor, McDonald, “What Is Disruptive Innovation?” https://hbr.org/2015/12/what-is-disruptive-innovation
Student tech presentations: GPS
Tuesday, October 17, 2023 – How to nurture innovation
Readings:
- Jon Gernter, The Idea Factory: Bell Labs and the Great Age of American Innovation (part one, chapters 1-11)
- Rosen, “Prologue and Changes in the Atmosphere”
Student tech presentations: birth control pill
Tuesday, October 24, 2023– Justice and innovation, 1
Readings:
- Benjamin, Race After Technology: Abolitionist Tools for the New Jim Code, 1-96.
- Lepore, “The Disruption Machine” https://www.newyorker.com/magazine/2014/06/23/the-disruption-machine
Student tech presentations: cement
Tuesday, October 31, 2023 – Justice and innovation, 2
Reading:
- Benjamin, Race After Technology: Abolitionist Tools for the New Jim Code, 97-end.
- Walter Benjamin, “The Work of Art in the Age of Mechanical Reproduction” (one copy)
Student tech presentations: antibiotics
Tuesday, November 7, 2023 – Beyond America and Europe
Reading: Digital Middle East, selections:
-
- Zayani, “Mapping the Digital Middle East: Trends and Disjunctions”, 1-32
- Any four (4) chapters of your choosing, based on your interests
Student tech presentation: eyeglasses
Three technologies to think about
Friday, November 10 – annotated bibliography due
Tuesday, November 14, 2023 – Critiquing technology
- Ursula LeGuin, “The Carrier Bag Theory of Fiction”
- Judy Wajcman, “Feminist Theories of Technology”
- Martin Heidegger, “The Question Concerning Technology”
- A fine and friendly podcast series about Heidegger’s big book
Student tech presentations: WiFi
Can We Contain Artificial Intelligence?: A Conversation with Mustafa Suleyman (Episode #332) – Sam Harris, Making Sense podcast https://youtu.be/IkojE37PUO8?si=Rsk7RGKAtmnlHofe
Debating the Future of AI: A Conversation with Marc Andreessen (Episode #324) [ same ] https://youtu.be/QMnH6KYNuWg?si=QhhaIQZNMn-sSrGV – This might be useful as an example of venture capital hubris.
On Model Collapse: “The Curse of Recursion: Training on Generated Data Makes Models Forget” –
https://doi.org/10.48550/arXiv.2305.17493 – Referenced in VentureBeat’s “The AI feedback loop: Researchers warn of ‘model collapse’ as AI trains on AI-generated content” https://venturebeat.com/ai/the-ai-feedback-loop-researchers-warn-of-model-collapse-as-ai-trains-on-ai-generated-content/
From a social justice perspective – a bit old in the current context but relevant as an implementation of AI within the surveillance domain: “Friction-Free Racism” – Chris Gilliard – https://reallifemag.com/friction-free-racism/
Excellent, Steven. Thank you for those. Good to see Chris Gilliard there.
I suggest:
Zuboff, Shoshana. The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power. New York, PublicAffairs, 2019.
Welcome to the Artificial Intelligence Incident Database. http://www.incidentdatabase.ai/.
Solarova, Sara, et al. “Reconsidering the Regulation of Facial Recognition in Public Spaces.” AI and Ethics, vol. 3, no. 2, May 2023, pp. 625–35. DOI.org (Crossref), http://www.doi.org/10.1007/s43681-022-00194-0.
Siemens, George, et al. “Human and Artificial Cognition.” Computers and Education: Artificial Intelligence, vol. 3, 2022, p. 100107. DOI.org (Crossref), https://doi.org/10.1016/j.caeai.2022.100107.
Saheb, Tahereh. “‘Ethically Contentious Aspects of Artificial Intelligence Surveillance: A Social Science Perspective.’” AI and Ethics, vol. 3, no. 2, May 2023, pp. 369–79. DOI.org (Crossref), http://www.doi.org/10.1007/s43681-022-00196-y.
Regan, Mitt, and Jovana Davidovic. “Just Preparation for War and AI-Enabled Weapons.” Frontiers in Big Data, vol. 6, May 2023, p. 1020107. DOI.org (Crossref), http://www.doi.org/10.3389/fdata.2023.1020107.
Mitchell, Melanie, and David C. Krakauer. “The Debate over Understanding in AI’s Large Language Models.” Proceedings of the National Academy of Sciences, vol. 120, no. 13, Mar. 2023, p. e2215907120. DOI.org (Crossref), https://doi.org/10.1073/pnas.2215907120.
How to Dance with and Intervene in Systems | Donella Meadows. http://www.hackernoon.com/donella-meadows-recommendations-for-how-to-dance-with-and-intervene-in-systems-92ace21743fb.
Heath, Alex. “Poe’s New Desktop App Lets You Use All the AI Chatbots in One Place.” The Verge, 28 Aug. 2023, https://www.theverge.com/2023/8/28/23849161/poe-ai-chatbot-platform-mac-app-llama-2-enterprise-tier.
Frankfurt, Harry G. On Bullshit. Princeton University Press, 2005.
Engelbart, Douglas. “Augmenting Human Intellect: A Conceptual Framework | Doug Engelbart Institute.” Doug Engelbart Institute, Oct. 1962, http://www.dougengelbart.org/content/view/138/.
Bostrom, Nick, and Eliezer Yudkowsky. “The Ethics of Artificial Intelligence.” The Cambridge Handbook of Artificial Intelligence, Cambridge University Press, 2014. Free access: http://faculty.smcm.edu/acjamieson/s13/artificialintelligence.pdf.
Bender, Emily M., et al. “On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? 🦜.” Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, Association for Computing Machinery, 2021, pp. 610–23. ACM Digital Library, http://www.doi.org/10.1145/3442188.3445922.
AI Fart Generator. http://www.boredhumans.com/farts.php. 😉
All from my AI & HE Resources doc: https://docs.google.com/document/d/1ELYfEXGrSvh2y6pKd6ne4093C_0uPUYmUWSJI0ltXNc/edit?usp=sharing
What a fine mix, Mark. Thank you.
I’ve used Bush’s “As we may think” in my big data/analytics course before (with undergrads) and it generates a lot of conversation and discussion.
I’ve also used the following pieces that may be of interest to you/this group…
Furman (2016). Is this time different? The opportunities and challenges of artificial intelligence. Remarks delivered at AI Now: The Social and Economic Implications of Artificial Intelligence in the Near Term. https://obamawhitehouse.archives.gov/sites/default/files/page/files/20160707_cea_ai_furman.pdf
Howard & Borenstein (2017) The Ugly Truth About Ourselves and Our Robot
Creations: The Problem of Bias and Social Inequity. https://doi.org/10.1007/s11948-017-9975-2
Rosenberg (2013). Data before the fact. In L. Gitelman (Ed.) , “Raw Data” is an Oxymoron. https://projects.iq.harvard.edu/files/eswg/files/rosenburg_-_rawdata.pdf or https://doi.org/10.7551/mitpress/9302.003.0003
Finally, I haven’t taught with this one yet, but I do like it a lot.
Filibeli (2019). Big data, artificial intelligence, and machine learning algorithms: A descriptive analysis of digital threats in the post-truth era. https://doi.org/10.16878/gsuilet.626260
Matthew, that’s a fine set. Thank you.
To add to a reading list:
A contrarian voice to AI regulation—Fear Wins: https://open.substack.com/pub/thealgorithmicbridge/p/fear-wins?utm_campaign=post&utm_medium=web
Warning Labels for AI text: https://clivethompson.medium.com/warning-labels-for-ai-generated-text-fa6235481631
A recent Mollick piece: https://www.oneusefulthing.org/p/what-people-ask-me-most-also-some?r=1ogu2y&utm_campaign=post&utm_medium=web
A useful reading I use with students:
Holmes, W., & Tuomi, I. (2022). State of the art and practice in AI in education. European Journal of Education, 57(4), 542-570.
This just out in a similar vein
Suchman, L. (2023). The uncontroversial ‘thingness’ of AI. Big Data & Society, 10(2), 20539517231206794.
Harry Frankfurt on Bullshit was mentioned in a comment abovw and there have been several pieces on this.
Mine is here, which is perhaps a trojan horse for me to talk about buddhism but at least is short!
Costello, E. (2023). ChatGPT and the Educational AI Chatter: Full of Bullshit or Trying to Tell Us Something?. Postdigital Science and Education, 1-6. https://link.springer.com/article/10.1007/s42438-023-00398-5
Great to see Ursula le Guinn and Carrier Bag fiction mentioned – will be coming back here to look more through your existing readings.
Eamon, that’s terrific, especially your own paper. Thank you so much.
I’m going to fire up a new post for this.
Rachel, thank you for those! Mollick is quite a generator now.
This sound incredible! The graph in the below website might pique curiosity or interest…
Maybe include the Gartner Hype cycle 2023 for emerging technologies.
https://www.gartner.com/en/articles/what-s-new-in-the-2023-gartner-hype-cycle-for-emerging-technologies
Good call, Roxann!
I’m going to recommend my own Ethics, Analytics, and the Duty of Care
MOOC content: https://ethics.mooc.ca
Shorter Paper: https://docs.google.com/document/d/1WM3u6ddxDQdK4FvxfEJEBlhdx0mJelgFj9YRJcyIZp8/edit
(I can give you a better document version on my website if you prefer)
Why this? It’s a comprehensive overview of the subject, without fearmongering or tech promotion, and avoids many of the straw man arguments and narratives we read today.
Stephen, thank you for this, which is new to me.
Where can I find that other version?
What a wonderful collection between the sources you’ve shared and all the others in the comments! Here are a few more:
A Blueprint For An AI Bill Of Rights For Education by Kathryn Conrad
https://criticalai.org/2023/07/17/a-blueprint-for-an-ai-bill-of-rights-for-education-kathryn-conrad/
What Happens When A Novice Writer Asks Chatgpt For Editing Advice? by Jane Rosenzweig
https://criticalai.org/2023/09/21/teaching-insights-what-happens-when-a-novice-writer-asks-chatgpt-for-editing-advice-dr-jane-rosenzweig/
Artificial Intelligence and Education: A Reading List by Anna Mills, JSTOR Daily
https://daily.jstor.org/artificial-intelligence-and-education-a-reading-list/
Now the Humanities Can Disrupt ‘AI’ by Lauren Goodlad and Samuel Baker
https://www.publicbooks.org/now-the-humanities-can-disrupt-ai/
How AI Reduces the World to Stereotypes by Victoria Turk
https://restofworld.org/2023/ai-image-stereotypes/
The New Laws of Robotics by Frank Pasquale
My dear Anna, thank you so much for these. And also for sharing your own trove at JSTOR!
There are some great links in the footnotes of my last blog post, including a link to Wolfram’s excellent “how does ChatGPT work” article.
https://darcynorman.net/2023/11/13/rethinking-about-generative-ai/#fn:1
What a fine post, D’Arcy. Thank you for putting it together and sharing.
Bryan,
This one is a bit more f a stretch, but I encountered it a day or two before you posted this. It comes by way of “The Syllabus” newsletter and was in their Essay of the Week feature, from 13 November entitled, “Balkan Cyberia” (https://lareviewofbooks.org/article/the-bulgarian-computers-global-reach-on-victor-petrovs-balkan-cyberia/).
In discussing Petrov’s book there is reference to a novel by a Bulgarian writer named Luben Delov. I have spent the morning here trying to track down the work (and ANYTHING in English by the author) and my lack of skill with Cyrillic is proving daunting.
According to the LARB essay, in 1981, Delov wrote a novel — and I quote the passage here —
“The most prodigious outgrowth of this cybernetic culture was science fiction, widely circulated and accessed through comics and book series. Petrov shares some of these, including fantastic premonitions of hallucinating AI bots, like The Missed Chance: Stories from My Computer (Propusnatiyat Shans, 1981), in which a computer linked to the national library’s databases produces stories for a bored writer until he realizes that it has been mockingly throwing his own writing at him all along.”
I did find some Bulgarian webpages referencing Delov, but gave up after trying to translate further (https://www.wikidata.uk-ua.nina.az/%D0%9B%D1%8E%D0%B1%D0%B5%D0%BD_%D0%94%D1%96%D0%BB%D0%BE%D0%B2.html). As horrible as that URL looks, it actually works.
~~ Cheers!
Dennis, that looks fascinating. I can’t think of the last Bulgarian sf I’ve read… although I did get called out once for missing a Bulgarian word in _Making Fiends_.
What an amazing list! Thanks for sharing this and reaching out for more ideas. Here are a few favorites:
A Blueprint For An AI Bill Of Rights For Education by Kathryn Conrad
https://criticalai.org/2023/07/17/a-blueprint-for-an-ai-bill-of-rights-for-education-kathryn-conrad/
An Introduction to Teaching with Text Generation Technologies
by Tim Laquintano, Carly Schnitzler, and Annette Vee, https://wac.colostate.edu/repository/collections/textgened/
What Happens When A Novice Writer Asks Chatgpt For Editing Advice? by Jane Rosenzweig
https://criticalai.org/2023/09/21/teaching-insights-what-happens-when-a-novice-writer-asks-chatgpt-for-editing-advice-dr-jane-rosenzweig/
Artificial Intelligence and Education: A Reading List by Anna Mills, JSTOR Daily
https://daily.jstor.org/artificial-intelligence-and-education-a-reading-list/
Now the Humanities Can Disrupt ‘AI’ by Lauren Goodlad and Samuel Baker
https://www.publicbooks.org/now-the-humanities-can-disrupt-ai/
How AI Reduces the World to Stereotypes by Victoria Turk
https://restofworld.org/2023/ai-image-stereotypes/
The New Laws of Robotics by Frank Pasquale
https://www.insidehighered.com/opinion/blogs/online-trending-now/2023/04/12/few-hidden-gems-among-ai-apps
I would take a more hands-on approach because as soon as we post these,
they are obsolete. Better, I think, to gain first-hand experience, and then back away and theorize that experience.
Reading about what someone else says is second-hand, or even third-hand.
I found the apps above to be invaluable for gaining direct experience both of limitations and un-limitations of AI.
That’s a good point about a lot of AI observation, Glen.
This is where a wiki would really come in handy as a living document.
This is how far professional accountants are using AI.
Not very.
There are 5 short (5min long) AI case applications at the bottom of this page.
I am not certain if you need membership or not.
Glen
https://www.cpa.com/gen-ai-practical-use-cases
No, it looks openly available. And some very good, practical examples of actual usage.
thanks for making this public, Bryan. Great crowdsourced moment. What about including some fiction? The Vara story “Ghosts” that was in the Atlantic (that I think is stunning) and this Sheila Heti piece: https://www.newyorker.com/magazine/2023/11/20/according-to-alice-fiction-sheila-heti
Terrific idea, Molly. “Alice” is amazing.
A professor of literature recommended Ted Chiang’s work to me. His 2010 “The Lifecycle of Software Objects” is available on Scribd.
https://www.scribd.com/document/333355838/The-Lifecycle-of-Software-Objects-by-Ted-Chiang
And that wonderful, tweeted a lot this weekend, visual article from the New Yorker. https://www.newyorker.com/humor/sketchbook/is-my-toddler-a-stochastic-parrot
Oh, excellent catch, Emma.
It’s impossible to keep up! This dropped last week:
Mitchell, Melanie, et al. Comparing Humans, GPT-4, and GPT-4V On Abstraction and Reasoning Tasks. Arxive, 13 Nov. 2023.
tl:dr – ” Our experimental results support the conclusion that neither version of GPT-4 has developed robust abstraction abilities at humanlike levels.”
Good one, Mark. That’s two papers with Mitchell as an author.
Mitchell, a fellow at the Santa Fe Institute, spoke on “The Future of Artificial Intelligence” in Santa Fe last week.
https://www.youtube.com/watch?v=GwHDAfAAKd4
Joe Lambert told me he was going to attend. I watched the livestream. We agreed my shorter presentation on AI to StoryCenter.org’s storytellers had better audience responses. 😉
https://www.youtube.com/watch?v=UH7eLhhuzvM