Read Generative AI and Creative Learning: Concerns, Opportunities, and Choices by Mitchel Resnick
As each new wave of technology ripples through society, we need to decide if and how to integrate the technology into our learning environments. That was true with personal computers, then with the internet, and now with generative AI technologies.

I just listened to the generated audio rather than read this.

Really powerful summary between the instructionist and constructionist approaches to AI in education. Resnick is of course the father of scratch, so is firmly on the constructionist side.

There are powerful ideas and examples of the ways AI could support a constructionist approach to learning and the 4Ps projects, passion, peers, and play.

I started to pull out quotes, but it easier to suggest you just read the whole thing.

 I worry that inertia and market pressures will push the educational uses of generative AI in this direction.

This would be the worry.

The piece finishes with

The choice is up to us. The choice is more educational and political than technological. What types of learning and education do we want for our children, our schools, and our society? All of us—as teachers, parents, school administrators, designers, developers, researchers, policymakers—need to consider our values and visions for learning and education, and make choices that align with our values and visions. It is up to us.

I do wonder if, in the mainstream, we have much choice. I don’t think that many decisions about educational technology have been very pure, the power of the big companies is massive. We should be thankful that the more open, non-commercial like scratch exists.

My class joined in the ‘ AI Wonderland: Unleash Creativity with Make it hAPPen (P4-P7)’ webinar on Monday. It was a useful introduction for their age group on a topic we had not explored in class. In Teams I noticed this TeachMeet1 too. I finally signed up for it on Wednesday.

Given it started at 3:30 on Thursday and school finished for the easter holidays at 2:30, it was a bit of a rush.

I had planned, the night before, to talk a bit about using ChatGPT for creating H5P content in Glow Blogs. I knocked up a quick keynote of screenshots to avoid the danger of live.

ChatGPT can quickly produce information which, once checked, can be used to create H5P content. What is especially useful is that it can format the information to work with HP5 textual inputs. I’ve put some instructions on the Glow Blogs H5P examples site.

The TeachMeet was quite quiet, 3:30 on the last day of term was probably tricky for most folk. I enjoyed the other things shared, although I didn’t grab any links, except for Diffit. I hope to get the rest when the recording is released.

Most of the sharing mirrored mine in that they involved creating resources, quizzes and the like. One idea that stood out, and one I intend to use, was taking an interesting phrase from pupils’ writing and using it as an image prompt in Bing (I believe). This was demonstrated to the whole class and sounds like it would generate interesting discussions.

I’ve used some of the free AI tools, mostly ChatGPT, for a while now. Mostly for simple text generation and some JavaScript or AppleScript help. I don’t doubt that, despite some glitches, that it is potential useful and interesting.

Is that an Elephant?

There are a lot of difficult and awkward questions around the use of GPT in teaching & learning. I’ve read a fair bit of discussion around the ethics at both ends of the process, but not much discussing the primary school level.

Things that worry me, beyond my knowledge, time, brain power or pay grade2:

  • The obvious, ethics around where the data comes from, scraping possible copyrighted works.
  • The bias of the data, racial bias is the one I’ve read about most, but I imagine there are many others.
  • Possible breaches of pupil/student data, safeguarding issues.
  • The commercial nature of the tools. A lot of these services seem to be freemium, with either a limited or time limited resource set.

I’ll keep using AI in a casual way with minimum risk (I hope), but it feels like education is stepping into a can of worms in the same sort of way we have adopted most technology, in a rather haphazard way.

Feature image is an old gif I made from a public domain photo a few years ago.

  1. I’ve not been to a TeachMeet for a while so this intrigued me as much as AI ↩︎
  2. Over the time it has taken me to type this post I see this: Women’s faces stolen for AI ads selling ED pills and praising Putin – The Washington Post, this AI – two reports reveal a massive enterprise pause over security and ethics. I also asked ChatGPT to give me 400 words on the pros and cons of using AI in education. I’d say there is a lot of confusion about. ↩︎


Yes, I can make a montage from image URLs. I can combine multiple images into one layout and add some effects and text to make it more appealing. 😊

To make a montage, you need to give me the URLs of the images you want to use, separated by commas. For example, you can type something like this:,,

can you make a montage from these images:,,,, mosaic style


Yes, I can make a montage from these images in mosaic style. I will try to make it interesting and colorful. 😊

Please wait a moment while I generate the montage for you. 🙏

A screenshot of 4 images generated by bing, not what was asked for.

Me: hmm I don’t see any of the images I asked to be used in the montage?

Bing: can you check … big list of things to check with my system…

no, you made a montage with some other graphics, not the images I specified.


I apologize for the confusion. I am not capable of creating montages or graphics. I can only describe images and provide information about them. If you could provide me with more information about the images you are referring to, I will do my best to help you. 😊

Is the Jetpack AI Assistant available for free?
Yes, the Jetpack AI Assistant block is currently available for free up to 20 requests. We encourage you to try it out and share your feedback. You can upgrade your plan to continue using the AI Assistant after the initial 20 requests.

I was quite interested in this, but not enough to pay £7.50 a month for it. The need for an account and payment also rules it out of Glow Blogs too.

I’ll continue to occasionally dabble with ChatGPT.

Bookmarked Degenerative AI in education by Ben Williamson.

But what if, instead of being generative of educational transformations, AI in education proves to be degenerative—deteriorating rather than improving classroom practices, educational relations and wider systems of schooling?

I’ve read this a couple of time, and probably need a couple more.

I guess, like other technologies that have entered the classroom, we are very much in the hands of the powerful. We get the technology we are given. Usually at a low cost, but perhaps at a high price?

As a side note, Jetpack now offers to change the tone of what I write with AI. Here is the humorous rewrite of the previous paragraph:

Continue reading


how could I display a list of links using the WordPress opml in the body of a page?

[opml url=”URL_OF_OPML-FILE”]

Replace “URL_OF_OPML_FILE” with the actual URL you copied in the previous step.
For example:

[opml url=””]

  1. Save or update the page.
  2. Preview or visit the page to see the list of links displayed.
    WordPress will parse the OPML file and generate the necessary HTML to display the links.

Are you sure there is an opml shortcode? This didn’t work on my site

I apologize for the confusion. You are correct that there is no built-in OPML shortcode in WordPress. I apologize for the incorrect information provided.

What was intriguing was the total confidence for the reply and the immediate apology, did it check to see if I was correct? It then went on to offer a plugin suggestion…

BBC Radio 4 – Word of Mouth, Chatbots 1

Like lots of other folk I’ve been reading plenty about Large Language Models, AI & Chatbots and playing with some of the toys.

I really liked Professor Bender’s approach and method. I also found this a very easy listen. My mind has tended to wander off when reading blogs post about AI. Very clear on the “not intelligent” and the risks associated with chatbots trained on large piles of language.

And specifically the things that they’re predicting is what would be a plausible next word given all the preceding words here and then again and then again and again.

And so that’s linguistically interesting that once you get to billions of words of text, there’s enough information in there just in the distribution of words to stick with things that are both grammatical and seemingly coherent.

So that’s a cool observation and it’s dangerous because we tend to react to grammatical, fluent, coherent, seeming text as authoritative and reliable and valuable.

So instead of talking about automatic speech recognition, I prefer to talk about automatic transcription because that describes what we’re using it for and doesn’t attribute any cognition to the system that is doing the task for us.2


  1. I subscribe to the RSS feed of this BBC radio program as a podcast, pity you can’t find the feed on the webpage.
  2. Ironically I used Aiko to get the text of the podcast for the quotes: “transcription is powered by OpenAI’s Whisper model running locally on your device”