80 Ways to Use ChatGPT in the Classroom (in Libby)
Since the release of ChatGPT in late 2022, the academic world is confronting the immediate reality of the ways in which artificial intelligence is changing the world, and the world of education in particular. While AI has been making inroads over the last few years, ChatGPT functioned as a wake up call for many. This guide is here to provide useful resources to understanding AI developments.
What are ChatGPT and its successor GPT4? They are generative pre-trained transformers.
ChatGPT and GPT4 (and there inevitable successors) are Large Language Models (LLMs). Having learned from billions of pages of web content, both good and bad, it produces text by applying rules it has generated from its billions of examples based on the problem it is being asked to solve.
Read TechScape: Seven top AI acronyms explained for a short article explaining the terminology of AI.
Generative AI tools can:
But consider Cassie Kozyrkov's important caveat: "ChatGPT is a bullshitter. The essence of bullshit is unconcern with truth. It’s not a liar because to be a liar, you must know the truth and intend to mislead. ChatGPT is indifferent to the truth." ("Introducing ChatGPT: The Revolutionary New Tool for Conversation Generation." Medium. )
As ChatGPT has become more omnipresent online, Ed Zitron's essay Are We Watching the Death of the Internet? looks at the morass AI is creating and the long-term risks for the internet itself.
DeepSeek, China's newly announced AI Chatbot is notable for a couple of reasons. It was built on cheaper chips, making a case for AI development not being dependent upon the most advanced technology as well as the speed with which the Chinese company caught up with American-based Nvadia. And second, it's open source model (a significant difference from Nvadia's proprietary model) "effectively show their train of thought and then use that for further training without having to feed them new sources of data" according to "What is DeepSeek, the Chinese AI Company Upending the Stock Market?" by Matt O'Brien for AP Business.
Stephen Wolfram explains, that at it's core, "what ChatGPT is always fundamentally trying to do is to produce a 'reasonable continuation' of whatever text it’s got so far, where by “reasonable” we mean 'what one might expect someone to write after seeing what people have written on billions of webpages, etc.'" It's working with probabilities, and adjusting the way it assesses "reasonableness" based on its interactions with the people setting it tasks. If you want a deeper understanding of how ChatGPT works, Stephen Wolfram offered a very detailed, yet understandable, explanation in his blog Writings on Feb. 14, 2023.
For less in depth understanding of what's happening (and importantly, what it's not doing), check out ChatGPT is Great, You're Just Using it Wrong.
You may also find it valuable to review the White House's Blueprint for an AI Bill of Rights, created by the Office of Science and Technology Policy. It defines "five principles that should guide the design, use, and deployment of automated systems to protect the American public in the age of artificial intelligence."
Here's a short video from CNN introducing the technology to viewers (1/24/23).
Here's 60 Minutes piece from 4/16/23.
If you like to listen to podcasts, here are some useful AI podcasts.