NowComment
2-Pane Combined
Comments:
Full Summaries Sorted

June 24 - July 1 AI Reflection Journal

Author: Kevin Hodgson

10 additions to document , most recent over 1 year ago

When Why
Jun-24-23 Saturday June 24
Jun-24-23 Looking Deep Into The Unknown
Jun-25-23 Sunday June 25
Jun-25-23 Robot Writing Help
Jun-26-23 Monday June 26
Jun-27-23 Tuesday June 27
Jun-28-23 Wednesday June 28
Jun-29-23 Thursday June 29
Jun-30-23 Friday June 30
Jul-01-23 Saturday July 1

Goals: exploring daily wonderings about the role of Generative AI for writing, making art, teaching and learning

DMU Timestamp: June 23, 2023 23:36

Added June 24, 2023 at 6:29pm by Kevin Hodgson
Title: Saturday June 24

My students ended their year yesterday and in final reflections, many of my sixth graders noted their appreciation for all the technology that was woven into our ELA lessons. That included some very basic explorations of Generative AI, and even then, I was wary about bringing it to the classroom. Some had heard of ChartGPT but many had not, and I wondered if I was opening a Pandora's Box for young students. In the end, I don't regret it. Bringing those discussions and explorations of ethics and AI advancements -- giving them information and time to process and ask question, and some basic explorations -- is what we teachers need to always be doing.

DMU Timestamp: June 23, 2023 23:36

Added June 24, 2023 at 6:30pm by Kevin Hodgson
Title: Looking Deep Into The Unknown

via Adobe Firefly

DMU Timestamp: June 23, 2023 23:36

Added June 25, 2023 at 5:35am by Kevin Hodgson
Title: Sunday June 25

I was reading a piece the other day (I can't find the link) about how all of the major Ed Tech companies are now gearing up for integrating AI into its products and marketing them to school districts. I suppose this was inevitable, but I still find it a bit worrisome -- this rush to product development before we've all had a chance to live with AI for a time and play with AI and learn with AI (which is why I appreciate this particular experiment, using the AI Thinking Partner concept). I am most curious about Google and its educational platforms, as we use our Google suite tools all the time. But I also saw that Google is not planning on AI integration into its educational suite of tools ... not yet, anyway. They are rolling it out to regular users first (probably to see how it is used and work out the kinks). It's difficult not think that in a few years, this AI integration will be just normal useage. I wonder how that will impact the art of writing?

DMU Timestamp: June 23, 2023 23:36

Added June 25, 2023 at 5:46am by Kevin Hodgson
Title: Robot Writing Help

via Dall-E

DMU Timestamp: June 23, 2023 23:36

Added June 26, 2023 at 5:44am by Kevin Hodgson
Title: Monday June 26

There was news last week that Paul McCartney was using an AI tool to complete a Beatles song featuring the mix of real and synthetic voice of John Lennon. The announcement caused a bit of an uproar, to be expected, given the passion of Beatles fans. McCartney defended the work, saying it was a worthy use of AI, to complete a work of art that would remain unfinished, given Lennon's early death. McCartney has long been one to experiment with music, and to use the recording studio as an instrument itself, so I was not surprised, and I will be curious to hear how it comes out. The AI impact on music is pretty fascinating, with the Drake/Weeknd song making headlines and then the Grimes experiment (using her voice AI tool to make tracks, as long as you give her credit and a cut of any sale profits). I've even done some inquiries and experiments with AI Audio platforms (the free ones, anyway) and although the results were mediocre, it is clear there is something there that could transform the music industry (for good? I don't yet know). It does seem to me, however, that AI voice is different than AI text, in that the human element of a person's voice/sound is more personal, and as we move towards AI video and AI audio, there are bound to be more ethical complications (who gives one permission to use someone's voice?).

Robot Beetle

DMU Timestamp: June 23, 2023 23:36

Added June 27, 2023 at 5:42am by Kevin Hodgson
Title: Tuesday June 27

I was reading an article in the newspaper about how Newark schools are piloting the use of AI in some classrooms, and how the results are mixed. And as with most things AI, there are voices on both sides of the issue coming through.

"Proponents contend that classroom chatbots could democratize the idea of tutoring by automatically customizing responses to students, allowing them to work on lessons at their own pace. Critics warn that the bots, which are trained on vast databases of texts, can fabricate plausible-sounding misinformation — making them a risky bet for schools." - via New York Times "In Classrooms, Teachers Put Ai Tutoring Bots To The Test"

Although the scale and use is different (run by Khan Academy, which eventually will sell its services), it reminded me of the experiments I am doing here, thanks to Paul, around thinking about Thinking Partners. Paul has offered to help me create a Thinking Partner for this space. I am curious but I feel as I don't know enough about the process to have deep thoughts on the matter.

I would like to figure out a Thinking Partner that has an art-centered approach to questions, so that I might engage it with queries around the art of making music and creativity. Maybe a jazz musician persona of some sort, a mentor with a wealth of knowledge.

jazz musician teacher

(created with Stable Diffusion)

DMU Timestamp: June 23, 2023 23:36

Added June 28, 2023 at 6:12am by Kevin Hodgson
Title: Wednesday June 28

Here's what I am wondering: Were any of my words and poems and stories and posts and art scraped to become part of the database that fuels sites like ChatGPT and Dall-E and others? I seem to remember coming across a site that would answer that question (although, how it would do it, I don't know) and I know there are some platforms that allow you to either opt out or block the AI Spiders from grabbing your content to feed its engines.

This conundrum of who owns what on the Internet is tangled, complex, fascinating and concerning -- all at the same time. (Are these words now getting fed into the machine?)

I've left a lot of words, in a lot of places, over the years. Sometimes, I wonder how I might gather them back and then think, that's impossible. Let it go (cue song from Frozen). But now I wonder if in some future point, there might be a way to use the AI database to cull back all of those lost poems and lost words. Maybe that's not likely.

Unlike some artists filing lawsuits, and for good reason, I don't feel all that put out if my words were used for Generative AI, and wouldn't it be fascinating to be confronted with some text that you lost but vaguely remember if it the AI output rang familiar when asking a question?

Eating Data

Image via Dall-E AI

DMU Timestamp: June 28, 2023 00:53

Added June 29, 2023 at 5:50am by Kevin Hodgson
Title: Thursday June 29

The Daily podcast, from the New York Times, had an interesting episode yesterday, in which they explored the imact of ChatGPT on colleges in the Spring 2023 semester.

They interviewed a history professor, who seems rather befuddled about how and why his students would use AI to generate responses to his discussion posts. He noted that he started to see something happening when the "voice" of some of his writers changed. That was his first indication that some students were just posting responses via ChatGPT.

The students interviewed had some interesting insights, too, about the ethical dilemmas they found themselves in, as this new technology appeared that could easily give them access to academic information. Some admitted using it. Others explained how they tapped into it as a sort of Thinking Partner. Most indicated that they knew it was probably wrong, and that they were paying good money to get a good education, so why find an easy path with AI?

The professor admitted that he did not speak to his classes about the role of ChatGPTand other AI in academic settings, even as his concerns mounted, but that he knows he has to. I think this is probably where a lot of educators at all levels are at -- wondering, what is this thing and how will it tranform my teaching and the way my students learn?

Ethical Dilemmas

AI Art by Craiyon

DMU Timestamp: June 28, 2023 12:38

Added June 30, 2023 at 5:30am by Kevin Hodgson
Title: Friday June 30

I don't why I keep reading about P(doom) but I do. P(doom) is the odds that someone gives to Artificial Intelligence causing the end of humanity. It's hard to put too much stock in this awful outcome but it's also difficult to just ignore it, too, particularly when it some of the AI scientists have P(doom) numbers hovering over 10 percent likelihood. I guess I remain fairly optmistic that the AI systems will not take that kind of turn for the worst (nuclear war, pandemic disease, etc.) but still, there are plenty of things to worry about in the here and now (misinformation, bias, etc.) when it comes to the Generative AI systems already settling into place.

If I had to settling on a P(doom) number, it might hover around 5 percent. Not likely to happen, but not out of the ballpark, either.

P(doom) Asteroid

DMU Timestamp: June 30, 2023 01:14

Added July 01, 2023 at 5:54am by Kevin Hodgson
Title: Saturday July 1

This is my last entry into this AI Dialogue Notebook Journal experiment and it has been helpful to continue my thinking and wondering and inquiry about Generative AI, and to use the AI Thinking Partner tools that are now embedded here in NowComment. With Paul's help and assistance, I have been working on creating my own Thinking Partner personas, and I know I need to spend more time with the Prompt Engineering resources that he and others have gathered.

I created three Thinking Partners now in play here:

  • JazzHands - a persona with a creative arts flavor
  • WriteOutRanger - a persona for Write Out that has a National Parks aspect (but I still need to do more work on it, if it is to be useful and used)
  • HaikuHere -- a poetry persona with a mission to scan a text and turn it into a Haiku poem

In a discussion with Paul, I started to realize that creating a Thinking Partner for reading a given text together (AI in conjunctioin with a real person) is different than creating a Thinking Partner to give feedback or do something with something the human has written (as I have been doing here all week). I need to think more on this, and how to fashion Thinking Partners in a way that make sense, but also doesn't just add more noise to a text.

David noted elsewhere, in his observations of what I was doing here, that I was using a sort of "shotgun" effect with the Thinking Partners as I was writing all week -- and that is true. I was not being very systematic with which AI I wanted to add and comment. I was more in exploration -- what will this one do? How is this response going to be different? What words and language were used to create this particular Thinking Partner and how does it impact the response?

Thinking Out Loud at the stars 91218

Created by Adobe Firefly AI

DMU Timestamp: June 30, 2023 01:14





Image
0 comments, 0 areas
add area
add comment
change display
Video
add comment

Quickstart: Commenting and Sharing

How to Comment
  • Click icons on the left to see existing comments.
  • Desktop/Laptop: double-click any text, highlight a section of an image, or add a comment while a video is playing to start a new conversation.
    Tablet/Phone: single click then click on the "Start One" link (look right or below).
  • Click "Reply" on a comment to join the conversation.
How to Share Documents
  1. "Upload" a new document.
  2. "Invite" others to it.

Logging in, please wait... Blue_on_grey_spinner