Hi everyone!
This issue is once again focused on generative AI.
First, I wanted to let you know about the four short tutorials I’ve created about ChatGPT and Generative AI.
We’ve released these at the U of Arizona Libraries, on our tutorials site.
(Total time to complete all four is about 35-40 min).
They contain short videos (3 min or less) and quiz questions for reviewing what you learned.
They are appropriate for students or instructors... or anyone new to this topic.
Feedback is welcome! They have a Creative Commons 4.0 license – so feel free to share with anyone.
And, as usual, this issue also includes:
Thought-provoking articles
“Just for fun”
Articles about the future.
I hope you’ll enjoy this newsletter and spread the word to your friends and colleagues.
️ ❤️ Enjoy! (And if you like this issue, please “heart” it - thanks!)
🤔 Thought-provoking
Five AI Research Tools That Reference Genuine Sources - Hong Kong University of Science & Technology
I believe it’s tools like these, based on generative AI models, that are the future of academic search (not ChatGPT itself).AI Attribution and Provenance - Tom Barrett
Some thoughts on going beyond traditional citations. “How do we best signal AI assistance when publishing without diminishing trust?”Accessibility Powered by AI: How Artificial Intelligence Can Help Universalize Access to Digital Content - The Scholarly Kitchen
A good overview of how AI can help with accessibility.Generative AI can be the academic assistant an underserved student needs - Sabrina Ortiz
”ChatGPT and other AI educational tools can increase accessibility and level the playing field for students of all backgrounds.”What The Beatles can teach us about AI - Bennet Institute for Public Policy
People and the press completely misunderstood how AI was used and so they freaked out.AI scores in the top percentile of creative thinking - Erik Guzik
and Automating creativity: There is now strong evidence that AI can help make us more innovative - Ethan Mollick
Two thought-provoking articles about creativity of (or with) generative AI.The Carbon Emissions of Writing and Illustrating Are Lower for AI than for Humans - Bill Tomlinson (UCal Irvine) et al.
It’s good when you hear stats about the climate impact of generative AI, to always ask, “compared to what?” Learn more in this talk by Koomey, “Misconceptions about IT energy & environmental impacts,” which is based on this paper, Does not compute: Avoiding pitfalls assessing the Internet’s energy and carbon impacts.
“A recurrent theme is that well-intentioned research often overestimates IT’s electricity use and climate impacts, sometimes by orders of magnitude. These results then become “factoids” that spread quickly as people share them and the media report them.”An Open Letter from Artists Using Generative AI - Creative Commons
Unlike the artists who are angry about generative AI models using their work for training, these artists are embracing and using generative AI. Read this summary article and the full letter to Congress from these artists.
”For us, generative AI tools are empowering and expressive. We use them not to duplicate others, but rather to make transformative new works and experiences.”Using AI to translate speech to multiple languages
(including synchronizing lips of the person speaking)
I used HeyGen translator for this experiment.
→ Watch me speak first in English, then Spanish, French, Hindi, and Chinese.
I’d love to hear from you if you speak any of those languages… let me know how accurate the translation is, thanks!
😄 Just for fun
How to Lo-Fi with AI - Nobody and the Computer
Fun YouTube channel about music creation with AI.
We watched the classic movie, Desk Set (1957) the other night and it reminded me of ChatGPT! So I made this little clip with annotations - just for fun. 😄
🤖 The future
This is the worst AI will ever be, so focused are educators on the present they can’t see the future - Donald Clark
”Those who do not see AI as a developing fast and exponentially, use their fixity of vision to criticise what has already been superseded. They poke fun at ChatGPT3.5 without having tried ChatGPT4, any plug-is or any of the other services available.”
Open challenges in LLM research - Chip Huyen
Useful summary of all the research areas that are being worked on, like reducing hallucinations, incorporating other data modalities, making LLMs faster and cheaper, designing a new model architecture, and more.
This is fascinating! I hope eventually though that an AI can help me learn better. I have a hard time learning and no one has enough patience to help. My problem is with Math. I can't even do basic math & I am 39. Nothing has helpe. I can learn anything else but also my memory is terrible. I need an AI that can help me with these issues.