General resources on ChatGPT and other generative AI tools
- Slides from Dr. Torrey Trust from University of Massachusetts Amherst: talks about what ChatGPT is, what it can do and can’t do (yet), and ideas for assessment design. Also has links to many other useful resources.
- The Sentient Syllabus Project, by Boris Steipe and others: includes suggestions for learning objectives, learning activities, and syllabus language in an era of generative AI. Includes three principles that faculty may wish to share with students: “(1) An AI cannot pass this course; (2) AI contributions must be attributed and true; (3) The use of AI tools should be open and documented.”
- AI Text Generators and Teaching Writing: Starting Points for Inquiry: a set of frequently-updated resources and links created by Anna Mills, hosted through the Writing Across the Curriculum Clearinghouse.
- Generative Artificial Intelligence Technologies and Teaching and Learning, from Monash University: suggestions for how to talk with students about generative AI, sample language for class guidelines around AI use, suggestions for assessment design, and more.
- Teaching Writing in the Age of AI, a series of five posts by Leon Furze on topics such as academic integrity, suggestions for assignment design for those that teach writing, and more. The link goes to the final post, which has links to the first four as well.
- Quick Start Guide to AI and Writing, a list of frequently-updated resources on various topics, from a Modern Language Association & Conference on College Composition and Communication joint task force on writing and AI.
How Large Language Models work
- Chiang, T. (2023, February 9). ChatGPT is a Blurry JPEG of the Web. The New Yorker.
- Edwards, B. (2023, April 6). Why ChatGPT and Bing Chat are so Good at Making Things Up. Ars Technica.
Ethical considerations around generative AI writing tools
- AI and the Future of Teaching and Learning, from Contact North: unpacks some of the larger ethical questions of using AI, among other topics.
- Teaching AI Ethics, by Leon Furze (January 2023): discusses several ethical issues related to generative AI tools.
- GPT-4 System Card from OpenAI (PDF): provides a list of potential ethical and other risks with GPT-4, and how the organization has attempted to address them. Note: examples of harmful content in inputs and outputs are provided in this document.
- Birhane, A. (2020). Algorithmic Colonization of Africa, Philosophy and Technology, 33(3), 491-500. doi: 10.1007/s13347-020-00417-9
- Bender, E.M., Gebru, T., McMillan-Major, A., and Shmitchell, S. (2021). On the Dangers of Stochastic Parrots: Can Language Models Be Too Big? In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency. Association for Computing Machinery. ISBN 9781450383097.
- Hovy, D., & Prabhumoye, S. (2021). Five sources of bias in natural language processing. Language and linguistics compass, 15(8), e12432. https://doi.org/10.1111/lnc3.12432
- Weidinger, L., Mellor, J., Rauh, M., Griffin, C., Uesato, J., Huang, P.-S., Cheng, M., Glaese, M., Balle, B., Kasirzadeh, A., Kenton, Z., Brown, S., Hawkins, W., Stepleton, T., Biles, C., Birhane, A., Haas, J., Rimell, L., Hendricks, L. A., … Gabriel, I. (2021). Ethical and social risks of harm from Language Models. ArXiv. https://doi.org/10.48550/arXiv.2112.04359
- Ferrera, E. (2023). Should ChatGPT be biased? Challenges and Risks of Bias in Large Language Models. Arxiv. https://doi.org/10.48550/arXiv.2304.03738
- Luccioni, S. (2023, April 12). The mounting human and environmental costs of generative AI. Ars Technica. https://arstechnica.com/gadgets/2023/04/generative-ai-is-cool-but-lets-not-forget-its-human-and-environmental-costs/
Assessment design and class guidelines
- A collection of 101 creative ideas to use AI in education: a crowdsourced open educational resource.
- How AUC Faculty are Addressing AI in Their Teaching Spring 2023: a list of example activities and assignments from faculty at the American University in Cairo.
- Mollick, E.R. and Mollick, L. (December 13, 2022). New Modes of Learning Enabled by AI Chatbots: Three Methods and Assignments. Available at SSRN.
This article discusses ways to use the output of AI Chatbots to improve student learning, including their ability to transfer knowledge and engage in critical evaluation. - Classroom policies for AI Generative Tools: a list of various class guidelines on generative AI use, including language used in syllabi. Curated by Lance Eaton at College Unbound.
Sample generative AI tools
Futurepedia has a very extensive list of AI tools, updated daily. We note here just a few of the many.
Generative AI writing tools
- Moonbeam: supports writing longer texts such as blog posts, articles, essays, and more.
- Sudowrite: creates first drafts based on concepts, users can ask it to describe, expand, rewrite, provide feedback, and more.
- Quillbot: provides a way to paraphrase and summarize inputted text, among other functions.
- Rytr: generates text according to templates such as blog posts, stories, emails; rewords, expands, shortens text
- Wordtune: generates text, rewords, paraphrases, expands text; also has a function that summarizes videos in YouTube
- Writesonic: generates text, summarizes articles, paraphrases, and more
- Search engines with chat functions, such as Bing AI Search Engine, perplexity.ai, and you.com.
Tools that provide support for research
- Elicit: can find relevant papers without perfect keyword match, summarize takeaways from the paper specific to a question, and extract key information from the papers
- NaimAI: helps with literature reviews by searching for articles according to keywords. Can also generate a literature review.
- Semantic Scholar: searches several publishers and preprint servers to provide academic papers relevant to a query, along with short TLDR summaries, abstracts, citations; also has a reader that provides tables of contents, highlights key points, and more
- Humata.ai: can summarize long documents such as PDFs, answer questions about them, and also support writing based on the content of those documents
Tools for generating code
- Github Copilot: suggests code and functions in real time based on context in comments and previous code.
- Tabnine: similar to Copilot, with a few differences including the claim to only train the models on open source code with licenses that permit such use.
Tools for generating images
- DALL-E 2: OpenAI’s image text to image bot; free for those who registered before April 6, 2023, otherwise paid
- Nightcafe: free for lower-resolution images; otherwise requires credits (some are provided free)
- Stable Diffusion: free image generator through a web interface that generates multiple images based on a single prompt
- Craiyon: free tier that takes longer to generate images; paid tier produces them faster
- Scribble Diffusion: generates images based on a user-inputted sketch plus text
Tools for generating presentations
- Tome: generates presentations based on text or even a short prompt, complete with AI-generated images
- Beautiful.ai: has a “designer bot” that can generate images or text or revise text for slides
Tools for creating or editing videos
- Kaiber.ai: AI-generated videos using AI-generated art, based on uploaded images or video, and music added by the user.
- Descript: edit videos by editing text transcripts, including cutting, removing filler words, changing words, and more
Other tools
- Goblin Tools: “a collection of small, simple, single-task tools, mostly designed to help neurodivergent people with tasks they find overwhelming or difficult” (from the About page)
