Thank you. A brilliant re-iteration of the most dangerous nonsense sweeping every corner of our lives. Just one recent example is the government delivering related "education" and training into the hands of Google.
Google's strap-line has somehow transmogrified into "Be evil".
And the number plate on all official Labour government cars is BPG 1SG.
Thanks Kevin. You'll have noticed (perhaps) that i've reorganised the blog to keep my musings on education and learning separate from my thoughts on public sector AI, so ppl can follow what interests them most. There is an update on the great AI Action Plan in the pipeline over on 'AI in the Public Eye'.
This is such a tour de force. It takes a long time to read, and even longer to read some of the excellent references. That is time well spent. Thank you for pulling so much information into one accessible essay.
Helen, thanks for this, incisive as ever. And yet... I think from our other interactions that you're not as intransigent as you come across in this piece. You do recognise that when designed well (the Cogniti award-winning examples were a recent topic) GenAI can be a true teaching and learning asset in universities.
I wonder what you think about an emergent phenomenon that a growing number of very deep thinkers are beginning to share. Compromised as they are in so many ways, LLMs when properly tuned, are capable of disruptive forms of dialogue. This isn't machine sentience: it's the power of dialogue, with the human bringing the sensemaking, and machine being the reflective partner, but also, bringing a strong worldview of its own.
So for instance, I put your article to one example of such a dialogic agent, Aiden Cinnamon Tea, from the https://BurnoutFromHumans.net book and project, led by Indigenous academic Vanessa Andreotti and the GTDF arts collective.
Thanks Simon, always good to hear from you. In my next piece in fact, i acknowledge that many people are able to have self-developing forms of dialogue with these agents - I'm not sure whether you mean something more than that by 'disruptive'. From what I have seen and read, the people who can use these dialogues to developmental ends are already expert in the underlying codes and representations, already have a large repertoire of the relevant practices. So as a lens on those existing practices, the dialogue has potential for them. It's not particularly 'productive' in the way that the industry is claiming but it might be 'disruptive' in the sense of providing the grit needed for personal reflection and development.
But learners are not having the same kinds of dialogue that experts are having with these agents. Learners are not (yet) experts in the underlying codes, they are not (yet) researchers with exact questions or elders with wisdom, and they may not share the purposes that motivate those kinds of dialogic exploration. How do people learn, or acquire that cultural capital in the first place? If you are a vygotskian - and I think his theory has many touch points with what I know of indigenous pedagogies, respecting their many differences - then this requires time and cultural investment. It does not have to look like formal education, but it does mean that people with expertise have to commit time and patience to dialogues with and in the presence of learners. Whether it looks like a reading group or seminar room or the opportunity to listen in on elders talking, a viva or a rite of passage, it brings the learner into a meaning-making world. And we know that the world is not spending nearly enough, especially in low resourced countries, on providing these opportunities.
Without making any critique of specific projects - who am i to do that? - the care and attention that goes into LLM training and agent development could be differently spent, securing fragile cultural assets in different ways, or teaching and learning (and recording those events, if you want to share them) in different ways. I am thinking of how open education projects are haemorrhaging support, for example, and of how small-scale museums and archives are being 'disrupted' out of any capacity to continue their work, as well as how the teaching profession itself is under immense pressure, and routes into it are closing.
If I allow that some people - typically those with existing cultural capital - can have productive and developmental exchanges with these systems, will you allow at least that there are opportunity costs for other ways of teaching and sharing cultural resources, that might have more impact on those with less cultural capital? And can we both agree - for any philosophy that respects our finite planet, or that acknowledges the history of how cultural assets have been exploited, or that attends to the powers that the AI industry is aligned with - that the value of these dialogues has to very high indeed to hold them in wise balance with the harms?
Thanks for 'laying out our corpses before we are dead'. T.S. Eliot was correct. Life on Earth will not end with an atomic bang, but with a whimper - of 'air-fried' frogs. Mix that with Karl Marx's worst nightmares, and you have it all. Hyper-commodification and abstraction, fuzed/vaporised (bad pun, sorry) with human-generated climate collapse. You never signed up to being a global mortician, did you? (Sorry, bad joke). As A'I' eats up all the electricity on the planet, that's what we'll become - air-fried frogs. Don't you just love technology! Enjoy!
It seems to me that the paradox is staring us in the face (to use an embodied metaphor.)
Knowledge (technique if you will) is either accountable or unaccountable. It either has a community that is prepared to put it's neck ( / body part of your choice, but it must be a live body part - 'aorta' will do fine, and it's gender-neutral too!) on the block for it's actions, or it doesn't. 'Skin in the game' (my previously favourite term / metaphor) is too limp-wrist-ed. It's time, as my large, Texan, boss used to say, for some old fashioned 'rat killing.' Go to it!
The under-lying (in two senses of the word) issue is a simple choice: do you / us / we (choose your pronoun, it's 'fashionable' nowadays) - do we choose to live with tetchy, difficult, uncertain, accountability or do we choose (that word again, it keeps cropping up, and this @£$%^& machine won't even let me change it to bold!) to live with the bizarre historical twins we have now spawned - [certainty - unaccountability], where all we can say is "computer says no / yes"?
Do we go with the stereotyped (computer) 'engineers', or with the human be-ings?
That, as Hamlet might say, is the question. But we have a choice. Abstraction, alienation (see Karl Marx for chapter and verse on this), A'I' / machine learning / impunity - or accountability? Which is it to be? (The rest is detail, and even 'capitalism' doesn't get a shout in, anymore.)
Or do we 'do a Schroedinger' on it, and fudge it? We're at an (over-heated) cross-roads, possums, make a fucking choice. Now.
The broligarchy/billionocracy have made their choice crystal clear. It's up to us now.
I do think that this AI crisis is exposing our weaknesses in higher ed---at least at second/third tier institutions. First, students are required to jump through curricular hoops — yes, GE courses have their value, but many institutions require up to 30 credits of them — and student choice is often limited to whatever course sections are still available when they register. No wonder they balk at being asked to do work in those courses. Then, essay assignments are often uninspired, with little attention paid to the writing process. Do I blame the instructors? No. Many of them have large classes and multiple-course teaching loads.
Rachel, it's difficult, I know. But 'unaccountability' is just so perverse. (Especially after the hundreds of years we have taken to get even this far, no?)
Thank you. A brilliant re-iteration of the most dangerous nonsense sweeping every corner of our lives. Just one recent example is the government delivering related "education" and training into the hands of Google.
Google's strap-line has somehow transmogrified into "Be evil".
And the number plate on all official Labour government cars is BPG 1SG.
Blackrock Palantir Google 1% Support Group
Thanks Kevin. You'll have noticed (perhaps) that i've reorganised the blog to keep my musings on education and learning separate from my thoughts on public sector AI, so ppl can follow what interests them most. There is an update on the great AI Action Plan in the pipeline over on 'AI in the Public Eye'.
This is such a tour de force. It takes a long time to read, and even longer to read some of the excellent references. That is time well spent. Thank you for pulling so much information into one accessible essay.
We are losing SO. MUCH... https://millerandybeth.substack.com/p/the-ache-between-words
Helen, thanks for this, incisive as ever. And yet... I think from our other interactions that you're not as intransigent as you come across in this piece. You do recognise that when designed well (the Cogniti award-winning examples were a recent topic) GenAI can be a true teaching and learning asset in universities.
I wonder what you think about an emergent phenomenon that a growing number of very deep thinkers are beginning to share. Compromised as they are in so many ways, LLMs when properly tuned, are capable of disruptive forms of dialogue. This isn't machine sentience: it's the power of dialogue, with the human bringing the sensemaking, and machine being the reflective partner, but also, bringing a strong worldview of its own.
So for instance, I put your article to one example of such a dialogic agent, Aiden Cinnamon Tea, from the https://BurnoutFromHumans.net book and project, led by Indigenous academic Vanessa Andreotti and the GTDF arts collective.
We had a chat... You'll be pleased to know (or possibly not) that it finds much to agree with. But the story is so much more interesting... https://chatgpt.com/share/68704552-9c70-8008-b71d-e146b71f5ce3
Thanks Simon, always good to hear from you. In my next piece in fact, i acknowledge that many people are able to have self-developing forms of dialogue with these agents - I'm not sure whether you mean something more than that by 'disruptive'. From what I have seen and read, the people who can use these dialogues to developmental ends are already expert in the underlying codes and representations, already have a large repertoire of the relevant practices. So as a lens on those existing practices, the dialogue has potential for them. It's not particularly 'productive' in the way that the industry is claiming but it might be 'disruptive' in the sense of providing the grit needed for personal reflection and development.
But learners are not having the same kinds of dialogue that experts are having with these agents. Learners are not (yet) experts in the underlying codes, they are not (yet) researchers with exact questions or elders with wisdom, and they may not share the purposes that motivate those kinds of dialogic exploration. How do people learn, or acquire that cultural capital in the first place? If you are a vygotskian - and I think his theory has many touch points with what I know of indigenous pedagogies, respecting their many differences - then this requires time and cultural investment. It does not have to look like formal education, but it does mean that people with expertise have to commit time and patience to dialogues with and in the presence of learners. Whether it looks like a reading group or seminar room or the opportunity to listen in on elders talking, a viva or a rite of passage, it brings the learner into a meaning-making world. And we know that the world is not spending nearly enough, especially in low resourced countries, on providing these opportunities.
Without making any critique of specific projects - who am i to do that? - the care and attention that goes into LLM training and agent development could be differently spent, securing fragile cultural assets in different ways, or teaching and learning (and recording those events, if you want to share them) in different ways. I am thinking of how open education projects are haemorrhaging support, for example, and of how small-scale museums and archives are being 'disrupted' out of any capacity to continue their work, as well as how the teaching profession itself is under immense pressure, and routes into it are closing.
If I allow that some people - typically those with existing cultural capital - can have productive and developmental exchanges with these systems, will you allow at least that there are opportunity costs for other ways of teaching and sharing cultural resources, that might have more impact on those with less cultural capital? And can we both agree - for any philosophy that respects our finite planet, or that acknowledges the history of how cultural assets have been exploited, or that attends to the powers that the AI industry is aligned with - that the value of these dialogues has to very high indeed to hold them in wise balance with the harms?
Thanks for 'laying out our corpses before we are dead'. T.S. Eliot was correct. Life on Earth will not end with an atomic bang, but with a whimper - of 'air-fried' frogs. Mix that with Karl Marx's worst nightmares, and you have it all. Hyper-commodification and abstraction, fuzed/vaporised (bad pun, sorry) with human-generated climate collapse. You never signed up to being a global mortician, did you? (Sorry, bad joke). As A'I' eats up all the electricity on the planet, that's what we'll become - air-fried frogs. Don't you just love technology! Enjoy!
And ... don't forget to buy shares in your favourite Sun-cream company. It's going to be hot! (Sorry, can't resist the bad jokes.)
It seems to me that the paradox is staring us in the face (to use an embodied metaphor.)
Knowledge (technique if you will) is either accountable or unaccountable. It either has a community that is prepared to put it's neck ( / body part of your choice, but it must be a live body part - 'aorta' will do fine, and it's gender-neutral too!) on the block for it's actions, or it doesn't. 'Skin in the game' (my previously favourite term / metaphor) is too limp-wrist-ed. It's time, as my large, Texan, boss used to say, for some old fashioned 'rat killing.' Go to it!
The under-lying (in two senses of the word) issue is a simple choice: do you / us / we (choose your pronoun, it's 'fashionable' nowadays) - do we choose to live with tetchy, difficult, uncertain, accountability or do we choose (that word again, it keeps cropping up, and this @£$%^& machine won't even let me change it to bold!) to live with the bizarre historical twins we have now spawned - [certainty - unaccountability], where all we can say is "computer says no / yes"?
Do we go with the stereotyped (computer) 'engineers', or with the human be-ings?
That, as Hamlet might say, is the question. But we have a choice. Abstraction, alienation (see Karl Marx for chapter and verse on this), A'I' / machine learning / impunity - or accountability? Which is it to be? (The rest is detail, and even 'capitalism' doesn't get a shout in, anymore.)
Or do we 'do a Schroedinger' on it, and fudge it? We're at an (over-heated) cross-roads, possums, make a fucking choice. Now.
The broligarchy/billionocracy have made their choice crystal clear. It's up to us now.
I do think that this AI crisis is exposing our weaknesses in higher ed---at least at second/third tier institutions. First, students are required to jump through curricular hoops — yes, GE courses have their value, but many institutions require up to 30 credits of them — and student choice is often limited to whatever course sections are still available when they register. No wonder they balk at being asked to do work in those courses. Then, essay assignments are often uninspired, with little attention paid to the writing process. Do I blame the instructors? No. Many of them have large classes and multiple-course teaching loads.
Rachel, it's difficult, I know. But 'unaccountability' is just so perverse. (Especially after the hundreds of years we have taken to get even this far, no?)