Bibs & bobs #14

 A wee rant

<BoR>

Maybe it was Marc Andreessen’s initial post on substack where he detailed how he would write.   

What’s my purpose? Variously: To share what I’m thinking; to share how I think; to share a way of thinking; to keep a personal notebook of thoughts; to send messages to my younger self.

… How will I write? Generally I won’t edit my copy, I won’t cite my sources, and I won’t try to be consistent. My motto is “strong views weakly held”, and you’ll see that on display here. Anything I say today I may disagree with tomorrow, in fact I frequently won’t even remember tomorrow…

What’s my hope? To show you that we live in a more interesting world than you might think; that it’s more comprehensible than you might fear; and that more things are possible than you might imagine.

Or maybe it was a webinar I watched this morning, but I felt the need for a smallish rant.

In writing this post, I can’t claim to be as freewheeling as Marc is but I do like the notion of strong ideas weakly held. If I’ve learned nothing these past years is that the play out of AI has caused me to rethink so many things that I used to hold as more or less reliable intellectual crutches. I’m having to learn to walk without them and make or find new ones even though I suspect they will be temporary.

I sat through an excellent Webinar given by Vitomir Kovanovic about the implications of AI for teaching and learning. Vitomir is part of the University of South Australia’s Centre for Change and Complexity in Learning (C3L). The centre has done much to draw attention to the emergence of AI apps and the implications for formal education. Importantly, Vitomir lightly mapped the history of developments in AI, an essential part of any understanding of the what, why and how of interplay in formal education. History matters. Understanding legacy effects, both material and intellectual matter. Without understandings of these patterns, any reaction to the new is blind. If you are going to think about or play in this space, these considerations are non-negotiables. You don’t need great detail but some sense of these ideas and history is crucial.

Webinars with chats can be difficult and annoying if you are trying to listen to a speaker and follow the always incoherent chat stream. I have a bad habit of reacting grumpily to comments that fall into the “you don’t get it” basket. I need to be more understanding and patient as the world of formal education slowly drags itself to deal sensibly with these early, in public terms, developments in AI. I should say that the video of the session was recorded as was the chat. 

I must also say though that I was pleased when on one slide Vitomir gave a number for the words used to train GPT3. His was 45 trillion. My back of an envelope calculation with help from ChatGPT was 48T. What’s a few trillion here or there in a LLM?

So with all the annoying chat going on I had to pause and remind myself that when new ways of doing things (aka technologies) emerge that “we” quite naturally apply familiar, comfy ways of thinking about things. That’s the first step in coming to terms with anything new. Sometimes the old ways of framing can be useful to a point. My experience, particularly around things digital is that they are not.

We still have folk who think in terms of an A4 world, i.e. a world in which print ruled and control of it shaped how we thought about things. Nicholas Negroponte coined the term “the digitally homeless” to describe such folk. 

Occasionally, you stumble on an idea that breaks from herd think and it adds usefully to your repertoire. Here, I am reminded of Jay Weston’s prescient paper [1]. It also brought to mind the time when automobiles emerged and people thought about them in terms of a change in transport, getting rid of horse dung from the streets and similar short-term consequences. It was an era of horseless carriage thinking. The other thing that came to mind was a book written by Carolyn Marvin: When Old Technologies Were New [2]. There is much to say about this fine piece of scholarship but a quote she took from a Henry Flad in the St. Louis Globe-Democrat in 1888 captures the idea well:

The time is not far distant when we will have wagons driving around with casks and jars of stored electricity, just as we have milk and bread wagons at present. The house of the future will be constructed with the view of containing electric apparatus for lighting, power, and cooking purposes. The arrangements will be of such a character that houses can be supplied with enough stored electricity to last twenty-four hours. All that the man with the cask will have to do will be to drive up to the back door, detach the cask left the day before, replace it with a new one, and then go to the next house and do likewise. This very thing will soon be taking place in St. Louis.

So that is my first point. We are as Avi Goldfarb puts it in “in-between times”, horseless carriage times. Choose your own analogy. 

What makes things as messy and as complicated as they are is well captured in a quote by William Gibson: 

The future is already here – it's just not evenly distributed.

The uneven distribution generates a great deal of noise, commentary and silliness that we are currently swimming in. It’s unhelpful but reflects strongly held views about what have been for a long time, the fundamentals of how we think about formal education.

Certainly we have to deal with the now, the availability of these new AI apps and what they mean for existing ways of doing things but unless we have an eye to the future we will be trapped unproductively in a horseless carriage world.

One of the positives in the chat this morning was a link to an eminently sane and sensible position on plagiarism [3]. Sarah Eaton’s offers six tenets. Anyone concerned about this issue needs to commit them to memory! While some might be seen as a bit of an over reach, to me the question for the longer term she draws attention to, is not if but when. 

All new ways of doing things are over-hyped in the short-term and under-estimated in the long-term and this is where the real problems lie for formal education. To me the developments longer term [4] ask huge questions about what it is to know, to learn, to demonstrate skill and knowledge, about what curriculum is, what it means to assess, what is worth knowing etc. These are the building blocks on which all of formal education’s current practices are built. To me they are all or will soon be fluid. Related to these matters are the old philosophical debates form the early days of AI in the 50’s and 60’s which have resurfaced. They remain important today.

A simple question I have asked for a long time, well before any of the recent developments in AI appeared: why do we teach students to do things that machines are good at? We wash clothes in a machine. We don’t use a copper to wash clothes as some of us will recall. We have a heavy reliance on an ever expanding set of machines both physical, digital and hybrid. The hybrids are physical machines with computers built into them.

In all of this we delegate work to machines. We pay little or no attention to what happens when we delegate. Here I indulge in one of my longer lasting intellectual crutches, that draws on notions from material semiotics more commonly referred to as actor-network theory. In very crude terms, Latour [5] and Sayes [6] perhaps more eloquently, argue that there is an exchange of capacities when we ask a machine to do work. The machine. in response, demands “new modes of action” from humans. 

To illustrate the point. At this time, I think that when you use ChatGPT to do a task you need to have three complementary skills/knowledge (new modes of action): you need to have a rough idea of how the app was built and how it works; you need to have good prompting skills; and you need to be able to evaluate what it produces. There are ample resources online that give good advice about the first two. I suspect these complementary skills will change over time as the AI systems improve, improve and improve again. I’m no fan of technological determinism. Nor do I think the social will control and manage things. It will all spin around the interplay between us and these new machines or apps. Yes they are black boxes. We can’t see inside them but that is also the case with the meat computer that sits on our shoulders. It is another, but different, black box although some of the logic that underpins the digital black box derives from the thinking about neurons and their associations. Much more to say. For now….

<EoR>

                                                                                                    

[1] Weston, J. (1997). Old Freedoms and New Technologies: The Evolution of Community Networking. The Information Society, 13(2), 195-201. https://doi.org/10.1080/019722497129214  

[2] Marvin, C. (1988). When Old Technologies Were New:  Thinking About Communications in the Late Nineteenth Century. Oxford University Press.  

[3] Posted by Sam Fowler in the chat. Attribution matters, even on small things. It is part of showing how you work and think.

[4] Just what is long and short in terms of systolic time is hard to make a call on. The developments in AI continue to attract massive investments which fuels the speed of development.

[5] Latour, B. (1992). Where are the missing masses? Sociology of a few mundane artifacts. In W. Bijker & J. Law (Eds.), Shaping Technology/Building Society: Studies in Sociological Change (pp. 225-258). MIT Press. http://www.bruno-latour.fr/sites/default/files/50-MISSING-MASSES-GB.pdf 

[6] Sayes, E. (2014). Actor–Network Theory and methodology: Just what does it mean to say that nonhumans have agency? Social Studies of Science, 44(1), 134-149. https://doi.org/10.1177/0306312713511867  

 

Bibs & bobs #13

 dis-ELIZAs

In the 1960s Joseph Weizenbaum created a natural language processing program called ELIZA. It was the first chatbot. It can be accessed here. ELIZA was designed to mimic a Rogerian counselor.  In broad terms the program did the prompting and the human user often replied at length which is the reverse of what we have now with chatbots like ChatGPT. While the response of the software was simple, their reaction to it prompted Weizenbaum to explore his ethical concerns in Computer Power and Human Reason [1].   

Having machines as talkative, responsive companions underpins the interest in having companion robots for the old and young. As Diana Fleischman argues, companionship extends to sexual gratification. 

A more intriguing take on this evolving pattern of interactions between humans and machines is provided by Kazuo Ishiguro’s book, Klara and the Sun [2]. It details a dystopian future in which socialisation of the young has been delegated to androids, AFs or artificial friends, humanoid robots powered by the sun.  The relation between humans and machines through the eyes of machines might seem far fetched at this time but for some time, humans have used machines to keep the young and old busy, interested and entertained. I’m thinking of radio, television, smart phones and, of course, games of all varieties. Ishiguro’s fiction explores a possible realigning of the human/machine relationship, to that of the machine.

It is easy to dismiss these musings for what they are, fictions or funny little imaginary worlds that form in a creaky old meat computer. But right now I think they are a largely unspoken part of the debate and commentary swirling around recent developments in generative AI. There is, for want of a better adjective, an emotional side to these debates.

Perhaps it will surface in interactions with various AI apps that slip into what Masahiro Mori calls the uncanny valley, that is the affinity we feel when interacting with an app becomes a repulsion. It poses an interesting question for the black belt prompters, how to develop prompts that produce that outcome. 

Maybe recent work by Carlos Perez will provide that reaction for you. 

Prospects for AI and research in the social sciences

I’ve always been interested in what I think of as secret academic business, i.e. how academics do their work: note making, searching, curation, writing and so on. Research across three European countries suggested that AI like a lot of the digital aids that have surfaced in recent years will go largely ignored.

94% of social scientists say they use Microsoft Word to write up their work. The vast majority also use Word to manage their references with just one in five using reference management software. Indeed, one in four humanities scholars draft early versions of their manuscripts in long hand. Most of the researchers we surveyed have no need for digital collaboration tools either – perhaps because the majority work alone – and where collaboration is needed, HSS scholars prefer to use email. They also overwhelmingly use email rather than social channels to promote their work. 

I need to locate more research that explores the attitudes and digital habits of researchers in the so-called soft sciences. There is obviously a visible, some might say noisy group who pick up on most of the shiny new toys that emerge from what remains a primordial digital soup. More broadly, and particularly in education, my sense is that there is a significant proportion of folk who get by with minimal engagement with things digital. There is much to tease out here. 

Persuasion

My curation of ideas is ordered around a small set of notions that I think are important for my making some sense of how the world works. Persuasion is one of those notions. Perhaps my engagement with the early actor-network (ANT) accounts of scientific practices underpins my interest. I found some of those ANT ideas in a review of Anand Giridharadas’ book The Persuaders by Ingrid Robeyns. 

My take-away from the book is that there is no point in believing you are right (or have the right policy, or the right analysis on what needs to happen on matter X), and believing the only thing that is needed for change is airing those views and that analysis. It’s just not enough. We need to actually spend time and effort to persuade others that this is the right analysis/policy/direction, and this persuasion cannot be merely cognitive; it requires understanding “where people are”, what makes them believe what they believe, and showing respect for them as a person at the outset. All of that requires listening, and being willing to engage in a genuine conversation, and finding out why people believe what they believe. Just believing I am right (and having all the arguments sorted out in my head) and airing my views, is not enough to also make a difference in the world, especially not in deeply divided societies.

While these ideas are often cited as ways to progress on issues of concern, you might have to conclude that not a lot of this approach is going on given the current status of differences across a wide range of issues. 

Of course all of this is easier said than done and does assume, as Robeyns points out that both sides of a debate need to have a commonality of how to play:

Being open to be persuaded implies being willing to accept, at a meta-level, that there is a distinction between a true claim and a false claim, and that one should not deliberately claim things that one knows to be blatantly false.

I found myself thinking about this analysis closer to home, as in how one persuades oneself and then to the roles machines play in persuasion which is well represented in commentary and analysis of advertising, recommendation systems and so on.  But I think there is something else going on.

It was perhaps timely that I came across a recent paper by Jeroen de Ridder [3] who makes an interesting case about what he called online illusions of understanding. He argues that,

the mere availability and use of internet search tools thus inflates people’s sense of how much they know and understand.  (p.5)

This got me thinking about the impact of prediction machines [4] on our illusions of understanding. One could argue that the machines we draw on are under no illusions about what they are, just ask ChatGPT. But in the hands of humans they can clearly contribute to the sense of inflation that de Ridder points to.  

Bottom line for the paper and for me is that all this adds up to a heightened importance of the ability to ask good questions. As I wrote that line I resisted the urge to revisit some of the lame stuff I have posed in this blog. But i think there is an important point here. Our largely A4-honed sense of what is a good question may be in need of some augmentation, augmentation that may even involve machines.

Value pluralism

An interview with Yejin Choi via Stephen Downes is a seriously useful read. Some snippets:

The truth is, what’s easy for machines can be hard for humans and vice versa. You’d be surprised how A.I. struggles with basic common sense. It’s crazy.

She draws on the notion of dark matter to talk about human common sense

Can you explain what “common sense” means in the context of teaching it to A.I.? A way of describing it is that common sense is the dark matter of intelligence. Normal matter is what we see, what we can interact with. We thought for a long time that that’s what was there in the physical world — and just that. It turns out that’s only 5 percent of the universe. Ninety-five percent is dark matter and dark energy, but it’s invisible and not directly measurable. We know it exists, because if it doesn’t, then the normal matter doesn’t make sense. So we know it’s there, and we know there’s a lot of it. We’re coming to that realization with common sense. It’s the unspoken, implicit knowledge that you and I have. It’s so obvious that we often don’t talk about it. For example, how many eyes does a horse have? Two. We don’t talk about it, but everyone knows it. We don’t know the exact fraction of knowledge that you and I have that we didn’t talk about — but still know — but my speculation is that there’s a lot. Let me give you another example: You and I know birds can fly, and we know penguins generally cannot. So A.I. researchers thought, we can code this up: Birds usually fly, except for penguins. But in fact, exceptions are the challenge for common-sense rules. Newborn baby birds cannot fly, birds covered in oil cannot fly, birds who are injured cannot fly, birds in a cage cannot fly. The point being, exceptions are not exceptional, and you and I can think of them even though nobody told us. It’s a fascinating capability, and it’s not so easy for A.I.

And her take on value pluralism

I’m excited about value pluralism, the fact that value is not singular. Another way to put it is that there’s no universal truth. A lot of people feel uncomfortable about this. As scientists, we’re trained to be very precise and strive for one truth. Now I’m thinking, well, there’s no universal truth — can birds fly or not? Or social and cultural norms: Is it OK to leave a closet door open? Some tidy person might think, always close it. I’m not tidy, so I might keep it open. But if the closet is temperature-controlled for some reason, then I will keep it closed; if the closet is in someone else’s house, I’ll probably behave. These rules basically cannot be written down as universal truths, because when applied in your context versus in my context, that truth will have to be bent. Moral rules: There must be some moral truth, you know? Don’t kill people, for example. But what if it’s a mercy killing? Then what?

ChatGPT, its initial and ongoing values training 

A good follow from the previous note is a detailed and clear account of ChatGPT’s training which teases out its cultural biases (ht David Weinberger). The more interesting light at the end of the tunnel is the growing army of folk who, by playing with it, contribute to its ongoing values training.   

The beat goes on

I find myself responding to various tweets that reflect an educational moral panic or yet another jaw dropping example of generative AI with the phrase: the beat goes on.

It thought it would be worth rewriting those lyrics to reflect on the impact of generative AI. So I prompted ChatGPT with: Rework these lyrics of the beat goes on to the impact of generative AI: and I supplied the lyrics.

It replied:

The AI beats on, the AI beats on

Models keep training, a dance to the brain

La de da de de, la de da de da

Generation was once the rage, uh huh

Technology has turned the page, uh huh

The neural network is the current thing, uh huh

Synthetic intelligence is our newborn king, uh huh

And the AI beats on, the AI beats on

Models keep training, a dance to the brain

La de da de de, la de da de da

The data sets are the supermart, uh huh

Machine learning still breaks its heart, uh huh

And researchers still keep on working hard

Electronically they keep the algorithms sharp

And the AI beats on, the AI beats on

Models keep training, a dance to the brain

La de da de de, la de da de da

Scientists sit and analyze

Bots keep learning to recognize

The progress keeps accelerating all the time

But humans still ask "Hey buddy, can you do this fine?"

And the AI beats on, the AI beats on

Models keep training, a dance to the brain

La de da de de, la de da de da

And the AI beats on (Yes, the AI beats on)

And the AI beats on (And the AI beats on, on, on, on, on...)

The AI beats on

If you need any evidence for the beat, this rather large collection of AI apps may do the trick.

                                                                                                    

[1] Weizenbaum, J. (1984). Computer Power and Human reason.  From Judgement to Calculation. Penguin.  

[2] Ishiguro, K. (2021). Klara and the sun (1^st ed.). Alfred A. Knopf.  

[3] De Ridder, J. (2022). Online Illusions of Understanding. Social Epistemology, 1-16. https://doi.org/10.1080/02691728.2022.2151331 

[4] See, Agrawal, A., Gans, J., & Goldfarb, A. (2022). Power and prediction : the disruptive economics of artificial intelligence. Harvard Business Review Press, and Agrawal, A., Gans, J., & Goldfarb, A. (2018). Prediction Machines: The Simple Economics of Artificial Intelligence. Harvard Business Review Press.  

Bibs & bobs #12

ChatGPT fun

Amid all the doom, panic and boosterism around ChatGPT there are tiny glimmers of fun. Here is one from Mark Schaefer: 20 Entertaining Uses of ChatGPT You Never Knew Were Possible. 

Accessible academic outputs

I mentioned the notion of embedding new text into LLMs which could then be interrogated. A model for this is Sahil Lavingia’s book, The minimalist entrepreneur. Here, you can ask his book questions. It’s roughly similar to Google’s talk to books. 

I began to wonder, what if academic publications all had an option like this. An option with which you could ask questions of the paper, chapter or book with the possibility of then extending your query via the references in the selected work. Yeah wild, but a fun shuffle of the academic game.

This would be a big step beyond what apps like Bearly offer currently. 

AI and writing

A collection of apps that use AI to support writing put together by Jeremy Caplan.

Transcription

There are a lot of options for transcripts of video and audio files. AssemblyAI is worth a play.  I was interested in a recent presentation by Venkatesh Rao to DEVCON in Bogota.  It mangled his name but the rest was pretty good. The app has a lot of options.

Formal Education and ChatGPT

As predictable as night follows day. Instead of doing the smart thing, i.e. given what machines can do some Oz universities will cling to pre-AI forms of assessment. So so so stupid.  The folk making these decisions are being paid big $. Who of them have explored or looked at how any of this is unfolding? Ban or domesticate is the old playbook formal education has used since the digital happened. It has not learned a thing from previous digital develops that go back to the late 1970s.  A measure of how smart any formal educational system or organisation is will be how well they deal with these developments, now and into the rapidly approaching future.  Below, I suggest the likely emergence of a new game that corporate universities will play: Whac-an-AI.

IMHO these developments will make Gutenberg seem like a tiny sneeze in the history of civilisation. 

Models, meddles and envy

The term physics envy is sometimes used to describe a motive behind some research in the sciences of the social. More broadly, the influence of different models drawn from science, i.e. Newtonian physics, chaos, non-linearity, complexity, emergent behaviour and so on, can be found in the logic that underpins a lot of research of the social and particularly in education which seems prone to picking up ideas that appear shiny or new, even if they are neither.

I’m of the view that there is nothing wrong with drawing on models and ideas from other fields but if you do, you need a decent helicopter view of the idea, its history and its limitations before drawing on it as metaphor or analogy.

For instance, if you wanted a quick and eloquent helicopter view of dimensions and the associated physics and mathematics it would be hard to go past a post of Margaret Wertheim’s. 

Perhaps it is an indication of how difficult it is to do good research in the so-called soft sciences that good helicopter views in this field of ideas, agendas and models are uncommon.  

Whac-an-AI

As each new bit of AI pops up to support writing, coding, planning and so on, and students studying in those fields continue to draw on them, you have a scenario for the perfect game of Whac-an-AI. In this game, the corporate university in order to protect it’s brand: our graduates don’t cheat, mistakenly commits to a practice of teaching their graduates how to do things that machines are now good at [1].

One thing that is predictable in this new wild space is that we will see more folk employed to manage and do the whacing. 

In anticipation of such a game emerging, I propose a new international standard for university stupidity called the Whac-an-AI.  Universities can be scored on an open ended scale by the number of moles they have more or less managed or tried to whac. Other awards spring to mind like best AI whac of the year, the most diligent AI whacer and so on. In time it could rival the ubiquitous ratings of universities that some universities appear to be obsessed with. If only :).

Sadly, if a tiny fraction of the effort associated with Whac-an-AI was put into:

1 supporting student understanding of what is going on in terms of LLMs and there many relatives

2 teaching students how to write half decent prompts and 

3 teaching students how to evaluate what LLMs output 

then life would be so much easier and better for students, staff and even managers!

The problem any university will face trying to ban or block AI use by students is that the moles will keep getting better and will breed rapidly. I am a bit of a fan of helicopter views of things. This tweet from Sterling Crispin puts the technical side of things into some perspective. 

Imagine an AI model that's 3x larger and more powerful than GPT3 aka ChatGPT

Google already built that in April, called PaLM, on their own TPU hardware competing with NVIDIA. People think ChatGPT will replace Google but they basically invented transformers in '17 (the T in GPT)

Imagine students playing with Monster moles!  The problem of course in an exponentially improving space is that the moles will keep getting better and will multiply faster than a corporate university could employ whacers.

Living through times that have elements that increase exponentially is an important part of making some sense of what is going on. As Dan Shipper recently wrote:

In his 1999 book The Age of Spiritual Machines, Ray Kurzweil wrote: “It is in the nature of exponential growth that events develop extremely slowly for extremely long periods of time, but as one glides through the knee of the curve, events erupt at an increasingly furious pace. And that is what we will experience as we enter the twenty-first century.”

A long time back I opted to use EdExEd as a label for my agenda. The Ex in the label is for exponentials. The other two: edges and education. Education should be plural I think.

                                                                                                    

[1] This is not a new phenomenon. Each new way of doing things has been met with attempts to ban or domesticate in formal education. A long time ago the hand held calculator was subject to banning and eventual domestication. Rarely was any thought given to the complementary knowledge and skills needed to make good use of the new way of doing things. There are lots of examples of this Dalek mindset.