ChatGPT is a blurry jpeg of the web

A really engaging and thoughtful explanation of the way large language models (LLMs) work, using surprisingly apt metaphors.

This analogy makes even more sense when we remember that a common technique used by lossy compression algorithms is interpolation—that is, estimating what’s missing by looking at what’s on either side of the gap. When an image program is displaying a photo and has to reconstruct a pixel that was lost during the compression process, it looks at the nearby pixels and calculates the average. This is what ChatGPT does when it’s prompted to describe, say, losing a sock in the dryer using the style of the Declaration of Independence: it is taking two points in “lexical space” and generating the text that would occupy the location between them. (“When in the Course of human events, it becomes necessary for one to separate his garments from their mates, in order to maintain the cleanliness and order thereof. . . .”) ChatGPT is so good at this form of interpolation that people find it entertaining: they’ve discovered a “blur” tool for paragraphs instead of photos, and are having a blast playing with it.

Nontechnical and a little bit long, but an extremely useful explanation for understanding how LLMs perform the magic they seem to accomplish. Could prompt some great discussions, especially among 11th-12th graders in a unit discussing AI technology.

(by Ted Chiang, author of several science fiction short story collections, including “Stories of Your Life and Others,” which included the story that the film “Arrival” was based on.)

Anatomy of an AI System

The Amazon Echo as an anatomical map of human labor, data and planetary resources

by Kate Crawford and Vladen Joler

A deeply fascinating, detailed, and critical examination of how a single product—the Amazon Echo—operates on many levels as an extractive technological object.

As we see repeated throughout the system, contemporary forms of artificial intelligence are not so artificial after all. We can speak of the hard physical labor of mine workers, and the repetitive factory labor on the assembly line, of the cybernetic labor in distribution centers and the cognitive sweatshops full of outsourced programmers around the world, of the low paid crowdsourced labor of Mechanical Turk workers, or the unpaid immaterial work of users. At every level contemporary technology is deeply rooted in and running on the exploitation of human bodies.

Might be a fascinating way for advanced students to begin considering what's lurking beneath the surface of AI technologies.

Kids Are Going Back to School. So Is ChatGPT

A good overview of the current capabilities of ChatGPT, possibilities, and concerns. Addresses a number of issues that are important for students to know about. This is a good piece for catching up on how the AI works and where its shortcomings lie.

MLA guidelines on citing generative AI

the MLA posted the following guidelines in March, 2023

You should:

• cite a generative AI tool whenever you paraphrase, quote, or incorporate into your own work any content (whether text, image, data, or other) that was created by it 
• acknowledge all functional uses of the tool (like editing your prose or translating words) in a note, your text, or another suitable location 
• take care to vet the secondary sources it cites (examples provided on the MLA site)

How students respond to allegations of using ChatGPT on assessments

This is the pre-press version of an article accepted for publication in Learning, by Dr. Tim Gorichanaz (Department of Information Science, College of Computing & Informatics, Drexel University).

Gorichanaz responses to ChatGPT accusations.pdf

US Naval Academy guidance

Shared with us by a Christchurch friend, here is the guidance for US Naval Academy instructors on creating classroom policies for the use of AI tools.

USNA Provost guidance on AI.pdf

Editorial argues against the use of ChatGPT in student work

[...] Using ChatGPT to complete assignments and write essays is worse than traditional plagiarism. Students come to college to get an education and become academically fulfilled. When students cheat, they are reducing their own capability to learn. Typing a prompt into GPT takes away the entire process of learning. Students who copy text from websites or books without citations are cheating, but at least they’re putting in some effort. [...]

ChatGPT theft laziness academic dishonesty.pdf

ChatGPT sparks debates about use in college classrooms

For example, faculty at Tufts University are split on whether ChatGPT should be used in the classroom. According to The Tufts Daily, some professors view usage of ChatGPT as academic dishonesty, while others see ChatGPT as an opportunity to revise syllabi in order to engage students in novel ways.

At UChicago and Beyond ChatGPT Sparks Debates.pdf

ChatGPT both prohibited and welcomed in classrooms

In his COMM305 syllabus, Senior Instructor Lance E. Schmeidler uses suggested syllabus policy language from the Stearns Center which may prohibit the use of AI. The Center gives recommendations to combat AI dependence such as clear communication of standards with students, adaptation of assignments as well as new grading criteria and emphasis on citations.
Schmeidler emphasizes that the use of ChatGPT should not be met with concern, and may even be encouraged in some courses. “The technology [AI] is still in its nascent stages and while intriguing, it is not yet a significant threat to the integrity of the learning experience. That said, unlike other forms of plagiarism that are less detectable, faculty have the advantage of being early-adopters and guiding students toward effective use of AI tools in their courses. We consider many negative consequences of adoption in the classroom, but over time education will evolve to include the technologies and elevate teaching and learning.” said Schmeidler.

ChatGPT both prohibited and welcomed.pdf

“Cheating service,” or powerful tool?

An early view that considers both sides.

“Each syllabus is supposed to specify what is considered academic honesty and dishonesty,” Mahoney said. “I'm sure you're going to have a sharp increase in cases where a professor is accusing a student of using this particular technology. I don't know how one goes about proving or disproving that.”

ChatGPT cheating service or powerful tool.pdf

Considerations when updating academic dishonesty policies

Students especially need to gain AI literacy in order to be competitive and effective in the job market, which is using AI more than ever. Yet all instructors must also develop strong AI literacy to be more relevant and effective in properly teaching these skills to students as a soft/power skill while also teaching the main subject of their instruction.

Is Using ChatGPT cheating plagiarism both neither.pdf

MLA guidelines on citing generative AI

the MLA posted the following guidelines in March, 2023

You should:

• cite a generative AI tool whenever you paraphrase, quote, or incorporate into your own work any content (whether text, image, data, or other) that was created by it 
• acknowledge all functional uses of the tool (like editing your prose or translating words) in a note, your text, or another suitable location 
• take care to vet the secondary sources it cites (examples provided on the MLA site)

MIT’s RAISE—Responsible AI for Social Empowerment and Education 

A ton of resources gathered in one place—from a wide variety of organizations—including professional development and learning for teachers as well as K-12 classroom units and projects. Some of these come from industry-supported sources, some from educational institutions, so there's a fair mix of approaches to AI including some good lessons on ethical concerns in AI (bias, fakes, etc.).

We’re not just interested in AI and education because it’s the “next new thing”… we want to make sure teachers and students are prepared to maximize AI’s potential and are also ready to help create lasting tech improvements for learners.

—SANJAY SARMA, Vice President, MIT Open Learning Initiative

Making sense of artificial intelligence

Offers “the A-Z’s of AI,” bite-sized explainers for a variety of terms and issues related to artificial intelligence. Fairly basic, but perhaps a good starting point for terms students might be unfamiliar with. Very slick presentation. Comes from an institute supported by Google, so it presents an overall positive picture here, without too much criticism.

The Elements of AI

A free online course: 

Our goal is to demystify AI

The Elements of AI is a series of free online courses created by MinnaLearn and the University of Helsinki. We want to encourage as broad a group of people as possible to learn what AI is, what can (and can’t) be done with AI, and how to start creating AI methods. The courses combine theory with practical exercises and can be completed at your own pace.

Part 1—Introduction to AI

An Introduction to AI is a free online course for everyone interested in learning what AI is, what is possible (and not possible) with AI, and how it affects our lives – with no complicated math or programming required.

Part 2—Building AI

Building AI is a free online course where you’ll learn about the actual algorithms that make creating AI methods possible. Some basic Python programming skills are recommended to get the most out of the course.

Don’t Want Students to Rely on ChatGPT? Have Them Use It

It’s easy to forget how little students and educators understand generative AI’s flaws. Once they actually try it out, they’ll see that it can’t replace them.

I decided to have each student in my religion studies class at Elon University use ChatGPT to generate an essay based on a prompt I gave them and then “grade” it. I had anticipated that many of the essays would have errors, but I did not expect that all of them would. Many students expressed shock and dismay upon learning the AI could fabricate bogus information, including page numbers for nonexistent books and articles. Some were confused, simultaneously awed and disappointed.

Draftback for Google Docs

Draftback is a browser extension for Chrome that allows you to play back a high-speed “movie” of all the edits to any document you have editing rights to. You can watch the document get written word by word and line by line, so anytime large blocks of text simply get pasted in it's obvious. Using Draftback is far easier than trying to look through the edit history of a student’s Google Doc by yourself and opening up all the triangles to see more detail on the changes.

Here's a 4-minute video showing how it works:

Creating “shared links” of an entire ChatGPT or Google Bard conversation

If you would like to help students use these tools more effectively and ethically, there is a way that students can share their entire chat history with an AI tool with you. If they create one of these links and share it with you, you can see the history of all the prompts they entered in a session, along with the AI agent's responses. It could be useful to require students to use these links when they create citations for their use of generative AI tools.

Above: Creating "Shared Links" in ChatGPT. Click on the three dots next to the chat listing, then click on "Share."
Link: ChatGPT help page explaining this.

Above: Creating "Public Links" in Google Bard. Click on the "share" icon at the bottom of the chat text, and choose "share" from the menu.
Link: Google Bard help page explaining this.

Chatbot Hype or Harm? Teens Push to Broaden A.I. Literacy

Teenagers have their own ideas about what they want to learn about A.I. But public schools rarely allow students to propel curriculum change or shape how they want to learn. That is what makes the student A.I. education campaign at River Dell High so unusual.

33 Free AI tools

List of free AI tools for a surprising number of applications, such as summarizing content, creating or analyzing music, etc.