This is the worst AI will ever be, so focused are educators on the present they can’t see the future

One thing many have not grasped about this current explosion of AI, it is that it is moving fast – very fast. Performance improvement is real, fast and often surprising. This is why we must be careful in fixating on what these models do at present. The phrase 'AI is the worst it will ever be' is relevant here. People, especially in ethical discussions, are often fixated by the past, old tools and the present, and not considering the future. It only took 66 years between the first flight and getting to the moon. Progress in AI will be much faster.

In addition to being the fasted adopted technology in the history of our species, it has another feature that many miss – it learns, adapts and adds features very, very quickly. You have to check in daily to keep up. 

Learning technology

The models learn, not just from unsupervised training on gargantuan amounts of data but also reinforcement learning by humans. LLMs reached escape velocity in functionality when the training set reached a certain size, there is still no end in sight yet. Developments such as synthetic data will take it further. This simple fact, that this is the first technology to ‘learn’ and learn fast, on scale, continuously, across a range of media and tasks, it what makes it extraordinary.

 

Teaching technology

There is also the misconception around the word ‘generative’, the assumption that all it does is create blocks of predictable text. Wrong. May of its best uses in learning are its ability to summarise, outline, provide guidance, support and many other pedagogic features that can be built into the software. This works and will mean tutors, teachers, teaching support, not taking support, coaches and many other services will emerge that aid both teaching and learning. They are being developed in their hundreds as we speak.

 

Additive technology

On top of all this is the blending of generative AI with plug-ins, where everything from Wikipedia to advanced mathematics, have been added to supplement its functionality. These are performance enhancers. Ashok Goes had blended his already successful teaching bot Jill Watson with ChatGPT to increases the efficacy of both. Aon top of this are APIs that give it even more potency. The reverse is also true, where Generative AI supplements other tools. There are no end of online tools that have added generative AI to make them more productive, as it need not be a standalone tool. 

Use and translation between hundreds of languages, also computer languages, even translation from text to computer languages, images, video, 3D characters, 3D worlds... it is astounding how fast this has happened, oiling productivity, communications, sharing and learning. Minority languages are no longer ignored.

All of the world's largest technology companies are now AI companies (all in US and China). The competitions is intense and drives things forward. This blistering pace means they are experimenting, innovating and involving us in that process. The prize of increased productivity, cheaper and faster learning, along with faster and better healthcare are already being seen, of you have the eyes to look.

People tend to fossilise their view of technology, their negativity means they don’t update their knowledge, experience and expectations. AI is largely Bayesian, it learns as it goes and it is not hanging around. People are profoundly non-Bayesian, they tend to rely on first impressions and stick with their fixed views through confirmation and negativity biases. They fear the future so stick to the present. 

 

Conclusion

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. It’s like using Wikipedia circa 2004 and saying ‘look, it got this wrong’. They poke the bear with prompts designed to flush out mistakes, like children trying to break a new toy. Worse they play the GIGO trick, garbage in: garbage out, then say ‘look it’s garbage’. 

This is the worst AI will ever be and its way better than most journalists, teachers and commentators think, so we are in for a shock. The real digital divide is now between those with curiosity and those that refuse to listen. Anyone with access to a smartphone, computer laptop or tablet... that's basically almost all learners in the developed world have access to this technology. The real divide is among those in the know and not in the know, using it and not using it, and that is the increasing gap between learners and teachers. So focused are educators on the present they can’t see the future. 

AI is now opening its eyes, like Frankenstein awakening to the world

The AI frenzy hasn’t lessened since OpenAI launched ChatGPT. The progress, widening functionality and competition has been relentless, with what sounds like the characters from a new children’s puppet show - Bing, Bard, Ernie and Claude. This brought Microsoft, Google, Baidu and Anthropic into the race, actually a two horse race, the US and China.

It has accelerated the shift from search to chat. But Google responded with Bard, the Chinese with Ernie’s impressive benchmarks and Claude has just entered the race with a 100k document limit and cheaper prices. They are all expanding their features but one particular thing did catch my eye and that was the integration of ‘Google Lens’ into Bard, from Google. Let’s focus on that for a moment.

 

Context matters

Large Language Models have focused on text input, as the dialogue or chat format works well with text prompting and text output. They are, after all, ‘language’ models but one of the weaknesses of such models is their lack of ‘context’. Which is why, when prompting, it is wise to describe the context within your prompt. It has no world model, doesn’t know anything about you or the real world in which you exist, your timelines, actions and so on. This means it has to guess your intent just from the words you use. What it lacks is a sense of the real world, to see what you see.

 

Seeing is believing

Suppose it could see what you see? Bard, in integrating Google Lens, has just opened up its eyes to the real world. You point your smartphone at something and it interprets what it thinks it sees. It is a visual search engine that can ID objects, animals, plants, landmarks, places and no end of other useful things. It can also capture text as it appears in the real world on menus, signs, posters, written notes; as well as translating that text. Its real time translation is one of its wonders. It will also execute actions, like dialling telephone numbers. Product search is also there from barcodes, which opens up advertising opportunities. It even has style matching.

More than meets the eye

OK, so large language models can now see and there’s more than meets the eye in that capability. This has huge long-term possibilities and consequences, as this input can be used to identify your intent in more detail. The fact that you are pointing your phone at something is a strong intent, that the object or place is of real, personal interest. That, with data about where you are, where you’ve been, even where you’re going, all fills out your intention.

 

This has huge implications for learning in biology, medicine, physics, chemistry, lab work, geography, geology, architecture, sports, the arts and any subject where visuals and real world context matters. It will know, to some degree, far more about your personal context, therefore intentions. Take one example, healthcare. With Google Lens one can see how skin, nails, eyes, retinas, eventually movements can be used to help diagnose medical problems. It has been used to fact check images, to see if they are, in fact, relevant to what is happening on the news.  One can clearly see it being useful in a lab or in the field, to help with learning through experiments or inquiry. Art objects, plants, rocks can all be identified. This is an input-output problem. The better the input, the better the output.

 

Performance support

Just as importantly, learning in the workplace is a contextualised event. AI can provide support and learning relevant to actual workplaces, airplanes, hospital wards, retail outlets, factories, alongside machines, in vehicles and offices - the actual places where work takes place - not abstract classrooms.

In the workplace, learning at the point of need for performance support can now see the machine, vehicle, place or object that is the subject of your need. Problems and needs are situated and so performance support, providing support at that moment of need, as pioneered by the likes of Bob Mosher and Alfred Remmits, can be contextualised. Workplace learning has long sought to solve this problem of context. We may well be moving towards solving this problem.

 

Moving from 2D to 3D virtual worlds

Moving into virtual world, my latest book, out later this year, argues that AI has accelerated the shift from 2D to 3D worlds for learning. Apple may not use the words ‘artificial’ or ‘intelligence’ but its new Vision Pro headset, which redefines computer interfaces, is packed full of the stuff, with eye, face and gesture tracking. Here the 3D world can be recognised by generative AI to give more relevant learning in context, real learning by doing. Again context will be provided.

 

Conclusion

Generative AI was launched as a text service but it quickly moved into media generation. It is now opening its eyes, like Frankenstein awakening to the world. There is often confusion around whether Frankenstein was the creator or created intelligence. With Generative AI, it is both, as we created the models but it is our culture and language that is the LLM. We are looking at ourselves, the hive mind in the model. Interestingly, if AI is to have a world view we may not want to feed it such a view, like LLMs, we may want it to create a world view from what it experiences. We are making steps towards that exciting, and slightly terrifying, future.

Huw did we ever get to this?

I spoke to an interesting woman at the BBC once, where I gave a talk on the challenge of digital media to traditional TV. My talk was received like a turd left behind by a burglar, as they then saw the internet and YouTube as an irrelevant gadfly. But that’s another story. At that event I met this woman, from Northern Ireland, who trained fledgling newsreaders and presenters. She told me she had informally called her course ‘The Egos Have Landed’ as she had repeatedly seen an odd phenomenon, young, and not so young journalists and others, catapulted into fame, thinking they were something more than autocue puppets. Their exposure turned them into monstrous narcissists who then started having opinions they thought mattered, all because they read from a teleprompter or chatted to each other on a studio sofa.

Saville was the King of such monsters, a prolific paedophile lauded, and worse, protected by BBC managers. Everyone knew, everyone laughed it off. Roll the credits on decades of paedophiles from Rolf Harris to Stuart Hall and a string of Radio 1 DJs. They’re an odd bunch. Kristian Digby, host of BBC1's To Buy Or Not To Buy, accidentally suffocated while attempting auto-erotic asphyxiation. We love a jolly frontman, as long as we don’t hear about his not so jolly backroom behaviour. Schofield and Edwards are just the latest in a long line of friendly faces that mask disturbing behaviour. I’m little concerned with their behaviour, as the witch hunts are so unedifying.

The deeper malaise is old media trying hard to avoid extinction. They need more front, as that’s the only thing they have left. Witness the recent disastrous interview by the BBC with Andrew Tate or Cathy Newman being demolished by Jordan Peterson. Whatever your views on these two odd chaps, they themselves have a lot of ‘front’, they’re smart, articulate and part of the counter-culture that has challenged TV. They ran rings around their stumbling, formulaic, ex-journalist interrogators.

The problem is too much focus on the ‘presentation’ layer. Presenters are really just juiced up human PowerPoints. I see this in tech all the time, its obsession with UX, then along comes Google – just type into a box, or ChatGPT, the same. TV has to put horrifically expensive lipstick on pigs because we want the truth watered down and mouthed out to us by what is known in the trade as ‘talking heads’. Loose Women, Quiz Shows and Reality TV are packed with these D-list ‘presenters’. They never die, just reappear as banal commentators on endless third rate entertainment programmes, the graveyards for clowns.

I have no idea why we think that news ’readers’ are worth listening to, outside of being working journalists. They’re the teleprompt and interview folk, and usually not very good at the latter, as their skills are with the written not spoken word. I was once introduced by Jackie Bird, a famous TV presenter in Scotland, as ‘Douglas’, even though I could see the word on the autocue was ‘Donald’. She was basically a bad parrot.

The problem is that they now get paid huge sums to ‘present’ homilies, seem like wholesome figures, often castigating others for their moral turpitude. We expect them to be our moral guardians, clean, pure, sensible and decent, when in truth they’re worse than most, as they often turn into overpaid narcissists. Will we miss Huw or will we manage without paying him £410,000 a year to read an autocue and behave like an old letch hunting down young ‘talent’?

I feel sorry for old Huw. He seems so ordinary, unremarkable and absent of charisma. Just a drone voice over royal events and a dull, earnest newsreaderI can't think of a single interesting sentence he ever uttered. He does stand out as someone without any obvious talent or presence.

TV is in trouble, as it is being crushed by the timeshifted streamers, social media and a dozen other alternatives. This is merely a sign of the old v new.

Is Ethics doing more HARM than GOOD in AI for learning?

I put this to an audience of Higher Education professionals at an Online Learning Conference yesterday at Leeds University.

I have an amazing piece of technology I’ve invented. It will bring astonishing levels of autonomy, freedom and excitement to billions of people. But here’s the downside, 1.4 million people will die horrible, bloody, sometimes mangled deaths every year, with another couple of million maimed and injured. This World War level of casualties, will strike every year, and is the price you have to pay. Would you say YES or NO?

Most rational souls would say NO. But let me reveal that technology – the automobile. We have come to an accommodation with the technology, as the benefits outweigh the downsides. Al may even bring in the self-driving car. My point is that we rush to judgement, as we are amateur ethicists and rely on gut feel, not reason.

This whole area, ethics, is oddly subject to a huge amount of bias as it is such an emotive subject. It plays to people's fears and prejudices, so objectivity is rare. Add new technology to the mix, along with a pile of stories in social media and you have a cocktail of wrong-headed certainty and exaggeration.

 

1. Deontological v Utilitarian

The offer I made at the start, I have put to many audiences. It is never taken up, as we are Utilitarians (calculating benefits against downsides) when it comes to actual decisions on using technology but dogmatic Deontologists (seeing morals as rules or moral laws) when it comes to thinking about ethics and technology. 

I am a fan of David Hume’s Indirect Utilitarianism, refined by Harsanyi as preference Utilitarianism. For a good discussion on how this relates to ethical issue and AI, see Stuart Russell’s excellent book, Chapter 9, Human Compatible (2019), where he attempts to translate this into achievable, controlled but effective AI. Curiously, Hume found himself cancelled by a few morally deluded students at the University of Edinburgh recently and they removed his name from the building which housed the Department of Philosophy. This was doubling down on Religious Deontologists refusing him a Professorship in the 18th century when he was one of the most respected intellectuals in the whole of Europe. Both groups are deluded Deontologists. He remains, in my opinion, the finest of the English speaking philosophers.This tension has existed in ethical thinking since the Enlightenment.

In truth, most of what passes for Ethics in AI these days is lazy ‘moralizing’, moral high horses ridden by people with absolute certainty about their own values and rules, as if they were God-given. More than this they want to impose those rules on others. They call themselves ‘ethicists’ but it is thinly disguised activism, as there is no real attempt to balance the debate out with the considerable benefits. It’s an odd form of moral philosophy that only considers the downsides. 

Google, Google Scholar, AI mediated timelines on almost all social media, the filtering out of harmful and pornography material into our email boxes, the protection of our bank accounts – all use AI. The future suggests that other huge near-term upsides in terms of learning, healthcare and productivity are well underway.

There is a big difference between ‘ethics’ and ‘moralising’. Even a  basic understanding of ethics will reveal the complexity of the subject. We have thousands of years of serious intellectual debate around deontological, rights-based, duty-based, utilitarian and other ways of thinking about ethics. A pity we give it so little thought before passing judgement.

2. Duplicity

Thomas Nagel points out, in his book 'Equality and Partiality', that we often pronounce strong deontological, moral opinions but rarely apply them in our own behaviour. We talk a lot about, say climate change, but drive large cars and fly off regularly on vacation. We talk about the climate emergency in academia but fly off for conferences at the drop of a sunhat, don’t deliver learning online and believe in spending €28 billion flying largely rich students around Europe through Erasmus. You may want all of your AI to be fully ‘transparent’. That’s fine, but stop using Google and Google Scholar and almost every other online service as they all use AI and it is far from transparent. My favourite example are those who are happy to 'probe' my unconscious in 'unconscious bias' training but decry the use of student data in learning on htebgrounds of privacy!

I’m just back from Senegal, where my fellow debating colleague Michael, from Kenya, berated the white saviours for denying the opportunities that AI offers Africa. Denying young aspiring workers to do human reinforcement training pays above the average wage and gives people a step into IT employment. It’s bizarre, he says, for white saviours on 80k to see this as exploitation.

3. To focus on AI is to focus on the wrong problem

Rather than climate change, the possibility of nuclear war, a demographic time bomb or increasing inequalities – AI is getting it in the neck, yet it may just solve some of these real and present problems. In particular, it may well increase productivity, democratise education and dramatically reduce the costs of healthcare. These are upsides that should not be thwarted by idle speculation.

At its most extreme, this speculation, that 'AI will lead to extinction of the human species' seems to have turned into the Doomsday tail that wags the black dog, despite the fact there is no evidence at all that this is possible or likely. Focus on what is likely not the fear-mongering that caught your attention on Twitter.

4. New technology always induces an exaggerated bout of ethical concern

Every man, women their uncle, aunt and dog, is an armchair ethicist but this is hardly new. It was ever thus. Plotus made the same point about the sundial in the 3^rd century: 

The gods confound the man who first found out how to distinguish hours! 
Confound him too who in this place set up a sundial to cut and hack my days so wretchedly

into small portions!

Plato thought writing would harm learning and memory in the Phaedrus, the Catholic Church fought the printing press (we still idiotically teach Latin in schools), travelling in trains at speed was going to kill us, rock ‘n roll spelled the end of civilisation, calculators would paralyse our ability to do arithmetic, Y@K was going to cause the world to implode, computer games would turn us into violent psychopaths, screen time would rot the brain, the internet, Wikipedia, smartphones, social media… now AI. 

As Stephen Pinker righty spotted a predictable combination of negativity and confirmation bias leads to a predictable reaction to any new technology. This inexorably leads to an over-egging of ethical issues as they confirm your initial bias.

 

5. Fake distractive ethics

Curiously, much of the language and examples in the layperson’s mind, has come from shallow and fake news, which is actually a real concern in AI, with deep fakes. Take the famous NYT article where the journalist claimed ChatGPT had told him to leave his wife. On further reading it shows he had prompted it towards this answer. If some stranger in a bar dropped you the line that his marriage was on the rocks, you’d put a significant bet on him being right to leave his wife. ChatGPT was actually on the money. It was a classic GIGO, Garbage In: Garbage Out, poke the bear story. Then there was that AI guided missile that supposedly returned back and hunted down its launcher - never happened – complete fake. The endless stream of clickbait ‘look it can’t do this’, mostly using ChatGPT 3.5 (a bit like using Wikipedia circa 2004), flooded social media. This is the worst AI will ever be but hey, let’s not consider the fact that first release technology almost always leads to dramatic improvement. Think long-term folks before using short-term clickbait to make judgements. 

 

6. Argument from authority

Then there is the argument from authority. I’m a Professor say people in strongly worded letter to the world, therefore I must be right. Two things matter here, domain experts often have a lousy track record and a lack of expertise in philosophy, moral philosophy, the history of technology, politics and economics. To be fair experts in AI are worth listening to as they understand what is often difficult to understand and opaque technology. Generative AI, in particular, is difficult to comprehend, in terms of both what it is, how it works and why it works. It confounds even AI experts. But they are not experts on politics, ethics or regulation.

 

The letters that appeared in both 2015 and 2023, pushed by Tegmark’s Future of Life Institute (whose role is ethical oversight), use the argument from authority. We’re academics, we know what’s right for you the masses. It demanded that we immediately stop releasing AI for six months until the regulators caught up – a ridiculous and naive request that showed their political, economic and social naivety. I dislike this ‘letter writing’ lobbying. First it had names of people who demanded they be taken off the list as they had not given permission and some have since rescinded the statement but authority alone is never enough.

 

Conclusion

This tsunami of shallow moralising is almost perfectly illustrated in Higher Education, where most of the debate around ethics has focused on plagiarism, when the actual problem is crap assessment. There is little consideration of the huge upsides and benefits for teachers and students alike. Learning, in my view is the biggest beneficiary of this new form of AI, health care second. Hundreds of millions are already using it to learn.

On climbing into personal pulpits, we may fail to realise the benefits in learning. Personalised learning, allowing any learner to learn anything, at any time from any place is becoming a reality. Functioning, endlessly patient tutors, that can teach any subject at any level in any language are on the horizon, universal teachers with a degree in any subject and driven by good learning science and pedagogy. The benefits for inclusion and accessibility are enormous, as its potential to teach in any language.

It is not that there are no ethical problems just that objective ethical debate is harmed when it becomes enveloped in a culture of absolute values and intolerant moralising. For every ethical problem that arises, there seems to be glib answers that are simple, confidently pronounced and often wrong.

I wrote this because I feel we are now in the position, in some countries and sectors, especially education, in getting bogged down in a swamp of amateur moralising on AI, suppressing the benefits. This has already happened in the EU, where the atmosphere is one of general negativity, seeing their role as regulators not creators. But the EU is only 5.7% of the world’s population. Google has not released Bard in the EU, OpenAI have set up shop in London and when Italy banned ChatGPT it spooked investors. We are in danger of throwing the baby out with the bathwater – and the bath. In practice learners are using this tech anyway, they are bypassing institutional inertia and high-horse ethical posing. Eric Atwell, at Leeds University, noted that all of his AI students ticked ‘not interested’ when it came to taking a module on ethics. They have a point. They know they’ll get a lot of moralising and not much in the way of ethics. It is unethical not to be using AI in learning.

Indeed, ethics may be doing more harm than good by making AI less useful and efficient. guardrailing and alignment may well be reducing the effectiveness of generative AI by placing too many constraints on output.

Leeds leads the way on HE and online learning events

Solid Online Learning Summit at Leeds University, open debate and discussion and some great people as speakers as well as expertise in the audience. I could only be there for one of the two days but it was worth the trip to Leeds, my second in a week. I like Leeds.

Irrepressible Neil Mosley 

First up, the irrepressible Neil Mosley, a knowledgeable, productive and honest broker of information on online learning in HE. Knows his stuff. He outlined the growth in the UK HE online learning market. I say growth but at 400k, in reality, it is a bit lacklustre, a point also made by both myself and the Paul Backsish. One could conclude that this is little more than a bit of an earner on the side, especially for foreign student income, rather than the strategic execution we see in the US. Taken by surprise by Covid, they seem to be retreating back into the old model and necessary expansion is slight. There is no real strategic intention to reduce costs and scale with online offers, as it is often an attempt to milk the lucrative 'Masters Degree' market.

His characterisation of the ‘partnerships’ market was good:

OPMs (Online Programme Management)

Ex-MOOC platform companies

Short Course Companies

Service Companies (learning design etc)

The whole MOOC movement made lots of mistakes and they’ve now turned into ‘courses’. The disaster that was Futurelearn, an organisation that simply ripped out cash from UK Universities, distracted them from the real task of online learning and collapsed as they had no business expertise. Hiring your CEO from BBC Radio condemned them to a long decline into irrelevance. The OU was meant to open up HE to a wider audience and could have led the charge into online learning but the old boys club took over and has thwarted them at evert turn.

Neil then looked at growth in the numbers and types of courses:

Degrees

MOOCs

Premium Short Courses

Micro-credentials

 

Micro-credentials

It is worth bringing in a later panel at this point on ‘Micro-credentials’, which must be one of the most disastrous bits of HE marketing ever… such a stupid word, an explicit recognition that what you offer is a trite piece of paper, badge or some such nonsense. It is such a stupid, demeaning and diminished term. The audience knew this but the panel seemed happy with it because it could be ‘translated’ – the worse response to any question on the day. This is what happens when you get people who know nothing about marketing talking about marketing. Not for the first time did the audience show real insights and expertise.

This rose by any other name stinks. A blatant attempt to, yet again, steal market share from those who do skills training well; FE and private providers. HE are hopeless at skills stuff but smell the cash and have been down lobbying the DfE, the panellist from Staffordshire admitted as much. The other panellist, from Wales, seemed to live on EU Erasmus grants, which have, rightly in my view, dried up. I did like the woman from Mexico who was blunt and honest about her very different context. Once again money gets sucked up from actual skills delivery to pretend skills delivery in HE. They can’t do this and justify this immoral move by tagging on the term ‘Lifelong Learning’. It doesn’t wash. HE is NOT in the Lifelong Learning sector, never was and never will be. There was also some baloney about ‘badges’ from a ‘badges’ man who we were told was some sort of lackey in the Royal Household. They will learn the hard way and fail to make money. Paul Bacsish made much the same point. I like Paul – he’s been around the block several times and has a good nose for this waste, which I remember him describing as ‘doomed to succeed’.

Learning Engineering

I enjoyed Aaron Kessler’s talk on Learning Engineering, although I’m not a fan of the term ‘engineering’ here as it is being used analogously. I feel that learning is a wide and messy business and doesn’t always fit neatly into this paradigm. The insistence of using learners in the process of design suffers, I think, from the obvious fact that they don’t know what they don’t know and are often delusional about good learning theory and practice. But the talk was sound, as it stated what is obvious, that process matters, implementation is hard and evaluation harder. The push towards data was also, rightly, emphasised. One again an audience member pointed out that most don’t have the luxury for the complexities of abstract model as they have tight deadlines (great point). Aaron very kindly gave me a copy of the ‘Learning Engineering Toolkit’ book, which has some pretty good stuff. I tackled the same stuff in my ‘Learning Experience Design’ book. We’re all in the same boat here, rowing in the same direction.

Ethics and AI

My contribution was a short talk on Ethics and AI. I made the point that most Ethical AI, is not ethics at all but ‘moralising’. It’s a complex issue diminished when barely disguises activism enters the room. Lots of moral high horses are being ridden into the debate, clouding expertise. The fact that HE focused almost entirely in plagiarism as the moral issue says how far behind we are in our thinking about the use of AI in HE. The problem is not AI but crap assessment. My message was a bit depressing as I now think the UK and EU are way behind on both AI and AI for learning. The US and China are streaking ahead as we wallow in bad regulation. Eric Atwell who teaches AI at Leeds very kindly summed my talk up by agreeing with every last thing I had said! This was gratifying as I find a great deal of good sense comes from practitioners, as opposed to arrivistes who have jumped on the ethical bandwagon. Adam Nosel made some good points about coaching and the need to maintain the human and social elements, as did Andrew Kirkton on some of the nitty gritty issues in HE.

Podcasts
I had breakfast with Bo from Warwick who was doing some great work on podcasting in her institution. It is a subject close to my heart. We are stuck in a traditional paradigm in learning design, ignoring one of the most important mediums of our day. Not to use podcasting in learning is mad, as hundreds of millions listen to learning podcasts every day, of their own volition. We know a lot about how to do these well and Bo was pn point here. Good to see young experts get a voice at this event.

Conclusion
These were merely my impressions written on the train back to Brighton, not an exhaustive summary and even though I disagreed with some, that is the point. Margaret Korosec Jo-Anne Murray, Megan Parsons and the rest of the team did a great job here, encouraging honest, open and sometimes uncomfortable debate. That’s the point. This is about moving forward, learning something new and moving on. To do that we need to look outwards. I’d have loved to have seen some people from FE here as well as private providers (there were some). But this was only the first event. It was a shame I couldn’t stay for the second day, the Tapas meal was fun, Leeds I love, and I met and spoke to some great people. Look forward to the second.

Can machines have empathy and other emotions?

Can machines have empathy and other emotions? Yann Lecun thinks they can and I agree but it is a qualified agreement. This will matter if AI it to become a Universal teacher and have the qualities of an expert teacher.

 

One must start with what emotions are. There has been a good deal of research on this, by Krathwohl, Damasio & Immordino-Yang, Lakoff, Panksepp. Also good work done on uncovering the role of emotion in learning by Nick Shackleton-Jones. l also covered them all in this podcast.

We must also make the distinction between:

Emotional recognition

Display of emotion

Feeling emotions

 

Emotional recognition

The face is a primary indicator of emotions and we look for changes in facial muscles, such as raised eyebrows, narrowed or widened eyes, smiles, frowns, or clenched jaw. Facial scanning can certainly identify emotions using this route. Eye contact is another, a solid gaze showing interest, even anger, while avoiding eye contact can indicate disinterest, shyness, unease or guilt. Microexpressions are also recognisable as expressing emotions. Note that all of this is often a weakness in humans, with a significant difference between men and women, also in those with autism. Emotional recognition is well on its way to being better than most humans and will most likely surpass that ability. 

 

Vocal tone and volume are also significant, tone of voice, intonation, pitch, raised volume when aroused or angry; quiet or softer tone when sad or reflective; upbeat when happy. Body language is another, clearly possible by scanning for folded arms and movements showing unease, disinterest or anger.

 

Even at the level of text, one can use sentiment analysis to spot a range of emotions, as emotions are encoded in laguage. LLMs show this quite dramatically. This can be used to semantically interpret text that reveals a whole range of emotions. It can be used over time, for example, to spot failing students who show negativity in a course. It can be used at an individual level or provide insights into social media monitoring, public opinion, customer feedback, brand perception, and other areas where understanding sentiment is valuable. As it improves, using LLMs it is starting to spot It may struggle with sarcasm, irony and complex language usage.

 

AI already could understands music in some sense, even its emotional intent and effect. Spotify already classify using these criteria using AI. This is not to say it feels emotion.

 

Even at the level of ‘recognition, it could very well be that machine help humans control and modulate bad emotions. I’m sure that feedback loops can calm people down and encourage emotional intelligence. The fact that machines could read stimuli quicker than us and respond quicker, may mean it is better at empathy than we could ever be. Recognising emotion will allow AI to respond appropriately to our needs and should not be dismissed. It can be used as a means to many ends, from education to mental healthcare. Chatbots are already being used to deliver CBT therapy. 

 

Display of emotion

Emotions can be displayed without being felt. Actors can do this, written words in a novel can do this and both can elicit strong human emotions. Coaches do this frequently. Machines can also do this. From the earliest Chatbots, such as ELIZA, that has been clear, Nass &Reeves, showed in 35 studies in The Media Equation, that this reading of human qualities and emotions into machines is common.

As Panksepp repeatedly says we have a tendency to think of emotions as human and therefore ‘good’. Their evolutionary development means they are there for different reasons than we think, which is why they often overwhelm us or have dangerous as well as beneficial consequences. Most crime is driven by emotional impulses such as unpredictable anger, especially violent and sexual crime. This would lead us to conclude that the display of positive emotions should be encouraged, bad ones designed out of the system. There are already efforts to build fairness, kindness, altruism and mercy into systems. It is not just a matter of having a full set of emotions, mort a matter of what emotions we want these systems to display or have.

 

Feeling emotions

This would require AI to be fully embedded in a physical nervous system that can feel in the sense that we feels emotions in the brain. It also seems to require consciousness of the feelings themselves. We could dismiss this as impossible but there are half way houses here and there is another possibility. Geoffry Hinton has posited The Mortal Computer and hybrid computer brain interfaces could very well blur this distinction in a sense of integrating thought with human emotions, in ways as yet not experiences, even subconsciously. But we may not need to go this far.

 

Are emotions necessary in teaching?

I have always been struck by Donald Norman’s argument “Empathy… sounds wonderful but the search for empathy is simply misled.” He argued that this call for empathy in design is wrong-headed and that “the concept is impossible, and even if possible, wrong”. There is no way you can put yourself into the heads of the hundreds, thousands, even tens and hundreds of thousands of learners. As Norman says “It sounds wonderful but the search for empathy is simply misled.” Not only is it not possible to understand individuals in this way, it is just not that useful. It is not empathy but data you need. Who are these people, what do they need to actually do and how can we help them. As people they will be hugely variable but what they need to know and do, in order to achieve a goal, is relatively stable. This has little to do with empathy and a lot to do with understanding and reason.

 

Sure, the emotional side of learning is important and people like Norman, have written and researched the subject extensively. Positive emotions help people learn (Um et al., 2012). Even negative emotions (D’Mello et al., 2014) can help people learn, stimulating attention and motivation, including mild stress (Vogel and Schwabe, 2016). We also know that emotions induce attention (Vuilleumier, 2005) and motivation that can be described as curiosity, where the novel or surprising can stimulate active interest (Oudeyer et al., 2016). In short, emotional events are remembered longer, more clearly and accurately than neutral events.

 

All too often we latch on to a noun in the learning world without thinking much about what it actually means, what experts in the field say about it and bandy it about as though it were a certain truth. But trying to induce emotion in the teaching and design process may not be not that relevant or pnly relevant to the degree that mimicing emotion may be enough. AI can be designed to induce and manipulate the learner towards positive emotions and not the emotions, identified by Panksepp and others, that harm learning, such as fear, anxiety and anger. We are in such a rush to include ‘emotion’ in design that we confuse emotion in learning process with emotion in the teacher and designer. It also seems like lazy signalling, for not doing the hard analysis up front, defaulting to the loose language of concern and sympathy.

 

Conclusion

In discussion emotions we tend to think of it as a uniquely human phenomenon. It is not. Animals clearly have emotions. This is not a case of human exceptionalism. In other words, beings with less complexity than us can feel. At what point therefore can the bottom up process create machine that can feel? We seem to be getting there and have come quite far having reached ‘recognition’ and ‘display; 

 

If developments in AI have taught us one thing, it is to never say never. Exponential advances are now being made and this will continue, with some of the largest companies with huge investments, along with a significant shift in research and government intentions. We already have the recognition and display of emotions. The feeling of emotions may be far off, unnecessary for many tasks, even teaching and learning.

 

In medicine, empathy is already being helped with GPT4, patients can benefit from being helped by both a knowledgeable and empathetic machine. We see this already Healthcare in the Ayers (2023) research, where 79% of the time, patients rated the chatbot significantly higher for both quality and empathy. That’s before the obvious benefits of being available 24/7, getting quicker results, increased availability of healthcare in rural areas, access by the poor and decreased workload for healthcare systems. It empowers the patient. For more on this area of AI helping patients with empathy listen to Peter Lee’s excellent podcast here. He shows that even pseudo-empathy can run deep and be used in many interaction with teachers, doctors, in retail and so on.

This is why I think the Universal Teacher and Universal Doctor are now on the horizon.

 

Bibliography

Ayers et al. 2023. Comparing Physician and Artificial Intelligence Chatbot Responses to Patient Questions Posted to a Public Social Media Platform

Norman, D.A., 2004. Emotional design: Why we love (or hate) everyday things. Basic Civitas Books.

Norman, D., 2019. Why I Don't Believe in Empathic Design.

Um, E., Plass, J.L., Hayward, E.O. and Homer, B.D., 2012. Emotional design in multimedia learning. Journal of educational psychology, 104(2), p.485.

D’Mello, S., Lehman, B., Pekrun, R. and Graesser, A., 2014. Confusion can be beneficial for learning. Learning and Instruction, 29, pp.153-170.

Vogel, S. and Schwabe, L., 2016. Learning and memory under stress: implications for the classroom. npj Science of Learning, 1(1), pp.1-10.

Vuilleumier, P., 2005. How brains beware: neural mechanisms of emotional attention. Trends in cognitive sciences, 9(12), pp.585-594.

Oudeyer, P.Y., Gottlieb, J. and Lopes, M., 2016. Intrinsic motivation, curiosity, and learning: Theory and applications in educational technologies. Progress in brain research, 229, pp.257-284.

https://greatmindsonlearning.libsyn.com/affective-learning-with-donald-clark 

Is the new Digital Divide in AI between the 'EU' and 'Rest of the World'?

OpenAI has opened its first foreign office in London citing the pro-innovation economy, talent and, it is clear although not stated, the fear of EU regulation. They are also clearly cautious about EU regulation. Bard was available in 180 countries and territories, including the UK, but NOT the EU, until a deal was done. Facebook has been holding back releases of models, Twitter has left the EU’s voluntary code of practice. Is this a new Digital Divide? One wonders what effect this will have on investment in AI across the EU? The lack of debate around the consequences of this is puzzling. 

When Italy declared UDI and banned ChatGPT they quickly relented (actually a move by a right wing appointee to show their strength). But this is different, large AI providers, such as Google, and Facebook, are taking the initiative and simply not releasing AI services in EU countries. This new Digital Divide may soon be between the EU and the rest of the world and could have serious consequences.

 

On the other hand, the EU is a huge and wealthy market, so the large tech companies will not take these decisions lightly. The problem is that the EU's legislation is often bureaucratic and cumbersome, involving lots of paperwork, hits on productivity. and is a stick and not carrot mechanism. The famous pop-up consent solution ‘Manage all cookies’ is GDPR nonsense, as no one reads the consent forms, it is therefore largely a waste of time. It was the result of bad legislation, producing a massive hit on productivity with no tangible benefits. One side overlegislates, the other is perhaps too lax and defensive – the net result is a Digital Divide.

With the release of Baidu's Ernie 3.5, which is neck to neck with ChatGPT4 on performance, this has turned into a two horse race - US and China. There's a third horse, but that's a moral high horse, which has barely left the stalls - that's the EU.

 

Economic environment

Let’s start with the big picture. 

“In 2008 the EU economy was larger than America’s. In 2008 the EU’s economy was nearly 10% larger than America’s at $16.2tn versus $14.7tn. By 2022, the US economy had grown to $25tn, whereas the EU and the UK together had only reached $19.8tn… Now the US is nearly one-third bigger. It is more than 50 per cent larger than the EU without the UK…” (FT 20 June 2023) and that gap is growing. 

The US has trounced Europe in terms of productivity, economic growth, investment models, research, investment, the creation of tech companies, defence and energy policy. It has also trounced the EU in terms of AI research and implementation. If the EU cannot develop a strong tech-based economy it will have to rely on low growth legacy markets, such as tourism and luxury goods, meaning it will fall further behind.

 

Productivity deficit

AI matters as ‘productivity’ needs a well-educated and skilled workforce, good infrastructure, and a favourable business and investment environment. Importantly, those with the more sophisticated tools tend to be the more productive. The evidence for the productivity gains using AI, within just a few months, is clear. If the EU either ban such tools or create an environment where angels fear to tread, then productivity on coding, management and general output, as these tools affect almost every sector, will start to lag. We had a dry run when Italy banned ChatGPT and there were reports of falls in productivity.

 

Training and education deficit

The University research and teaching system that feeds AI tech in terms of core research and skilled labour is dominated by the US, UK and China. EU Universities barely figure in the major rankings. An additional problem is the now deeply rooted anti-corporate sentiment in Higher Education in the EU. The sneering attitude towards the private sector, even OpenAI as a not-for profit, is now the norm, often accompanied by a failure to understand its actual structure. A symptom of this is that the debate in Higher Education focused largely, not on learning, but plagiarism. Far too little debate has taken place on the benefits in education and health.

 

Effort in AI is skewed towards often vague ethical initiatives making the overall atmosphere one of negativity, slowing down progress. The danger is that the benefits will be realised elsewhere while the EU remains rooted in old, analogue instititutions, where everything in tech is seen as a moral problem. There is nothing wrong with the moral debate but it is so often driven by fearmongering and activism, and not objective moral debate, which is to look at the moral issues and consequences, good and bad, not just the bad.

 

An additional problem is the unlikely adoption of AI in education in the EU. The real initiatives such as Khan Academy and Duolingo have been funded and implemented in the US, aided by philanthropic investment. There is little of that energy and type of investment in Europe. As AI becomes integrated into education and training in the US, its absence here will mean less productivity. 

 

Investment freeze

Investors looked askance at Italy’s surprise ban and widened their astonished gaze across the whole of the EU. If one country can do this, so can others. Investors have a currency – it is called ‘risk’. They assess and quantify risks and base decisions based on that risk analysis. Anyone who has been through the process knows that they do their homework and due diligence. One of those risks is already baked in, the Italy ChatGPT ban, huge punishment fines is another, the generally negative rhetoric and cultural context is yet another. Why would large scale investors pump cash into a territory where bans, fines and an absence of services have become the norm? Investors like a favourable business environment not one that is based on negativity and punishment.

 

Investment model

The model that emerged post-war in the US has proved superior to that in the EU – private sector, investors, government and Universities working together on large projects with a real focus on impact. In AI we now see the fruits of that system in the US, where ground-breaking research on foundation models take place in large tech companies and not-for-profits, such as OpenAI. Europe scoffs, and gets bogged down in long-winded, bureaucratic and low impact Horizon projects, while the US and China gets on with getting things done. In truth we now look to the US for investment and that is the market most want to expand in, as it is the largest growth market on the planet. They have become so dominant that they merely buy European companies in AI.

 

Ethical quicksand

Generative AI has launched a thousand quangos, groups and bad PhDs on ‘AI

and ethics’ across Europe. You can’t move for reports, frameworks and ideas for regulations which rain down on us from publicly funded organisations, with far too little attention on potential solutions and benefits. 

 

Debate on the benefits has been swept aside by pontificating and grand-standing. It is easy to stand on the sidelines as part of the jeering, pessimist mob, less easy to do something positive to actually solve these issues. Rather than solve the problems of safety, security, alignment and guard-railing with real solutions, the EU has chosen to see the glass, not as half full, but as brimming with hemlock. It sees laws and fines as the solution, not design and engineering.

 

The EU also has no laws banning VPNs and their use is becoming more common. This is a huge loophole when using AI services. It is already happening with Bard, as the internet is like water, it tends to seep round and into places, based on demand.

 

Punishment strategy

The EU has been issuing fines for some time now, although not is all as it appears. You may note that many of these large fines are issued from Ireland, but there has been a long and bitter fight between Dublin and Brussels. Ireland gains a good portion of its GDP from a small number of US tech companies and because they are based there, the GDPR fines come from there. They have fought these fines tooth and nail but, in the end, had to bend the knee to Brussels.

 

There is something reasonable in these latest fines as the data transferred may be used by US surveillance agencies (they have a bad track record here). In practice Meta have until later in the year to comply. This is a bit of a cat and mouse game, with politics at the heart of it all. On the other had the EU puts up with Ireland and Luxembourg stealing other countries tax revenues through massive tax evasion. It is all a bit of a tangled mess.

 

The bottom line, however, is that this tactic is resulting in a deeper rift. Twitter quit the EU’s voluntary code of practice in mid-May as it could be fined up to 6% of its global revenue (£145m) or be banned across the EU if it does not comply with the Digital Services Act. Facebook are making noises about abandoning the EU.

 

Solutions

If as much effort went into solutions, than regulations, fines and rhetoric, we would progress at the right pace, solving problems as we go, rather than trying to punish people into submission. Hacker-led safety testing, well funded research, effort on international ISO standards not regional efforts with a focus only on large operation and parameter models and implementations would all help. Above all third-party professional hacker teams can be deployed to identify security and data weaknesses before release. Incident reporting can also be useful. This collaborative, non-confrontational approach is far preferable to the negativity and sledgehammer of legislation and punishing fines.

 

Conclusion

These battles have been raging for some time, mostly behind the scenes but Google and Facebook have also had run ins with Canada and Australia. Some of this has been resolved, some not. There is something predictable about it all - the old world versus the new.

It is an inconvenient truth but the EU is too late to the party, the US and China have forged ahead in IT and AI with their own tech giants. The EU has failed to create tech giants and has also deliberately chosen the path of being some sort of global regulator but it has a weak economy, weak research, weak investment and a weak entrepreneurial culture. In the same way that the Ukraine war showed the EUs lack of investment in defence and a lack of any overall defence policy, where not for the first time it had to rely on the US to come to its aid and provide arms, cash and expertise, so it is with AI. 

 

The investment is low, there is no policy other than taking a morally superior stance. So much energy has gone into ethical hand-wringing that Europe is reduced to being a bystander. It thinks it has sway but it is a fraction of the world’s population, shrinking, and white Eurocentrism now seems more than a little dated. It rides its lumbering moral high-horse, looking down on the rest of the world, while others like the US feed it a little hay to keep it happy and speed past. It would surely be better developing AI solutions that have identified benefits in productivity, learning and healthcare, than simply regulating it.

Pask, Conversational Theory & Generative AI

An often forgotten learning theorist, who needs to be revisited in the light of Generative, conversational AI, is Gordon Pask (1928 – 96). His work on learning machines and conversational theory, although reaching back to the 1950s, has turned out to be both prophetic and useful. 

Known as the ‘Dandy of Cybernetics’ he was famously mannered, theatrical, intense and eccentric. With his cape, Edwardian suit, bow-tie and pipe, he was known as being difficult to communicate with, his lectures and writing often difficult to comprehend. Both an academic and entrepreneur (largely unsuccessful) he liked the freedom to build and experiment and valued his autonomy. 

 

Nevertheless, he was an original thinker who has made a significant contribution to thinking about the complexity of learning. A polymath interested in geology, engineering, art, theatre, sculpture, biological computing, artificial intelligence, cognitive science, logic, linguistics, psychology music, cybernetics and education, he saw learning as a complex process of interaction.

 

Well aware of Artificial Intelligence, he was a friend of Marvin Minsky and would stay with him in the US, although he was critical of the AI community’s tendency to focus on isolated systems, separate from human interaction. As AI has widened out into language models and become part of the global system of learning, Pask has come back into focus, with his sophisticated theories of learning through conversation.

 

Learning machines

Like Pressey and Skinner before him, he developed a number of learning machines. The difference was in seeing machines as things we converse with, rather than use as tools. In this sense he was closer to Pressey than Skinner. Eucrates, for example, not only taught but could adapt its teaching based on knowledge of the learner. There is physical feedback, and the machine builds a model of user behavior. He drew upon Wittgenstein’s idea of meaning as use. This is in stark contrast to Skinner’s behaviourist reward machines.

 

Developed in the 1953s, he invented Musicolour, a system of interaction between performer and system that allowed a light show to be triggered by a musician. It could amplify a performance, even become ‘bored’ by the musician’s repetition! It went on tour around Music Halls in England and was interaction as performance.

 

A more practical learning machine was SAKI (Self-Adaptive Keyboard Instructor), built in 1956, to teach punch-card operators, with electric wires that cued the learner, detected keypresses and time intervals, always pushing the learner ahead but not too far ahead, to keep the learner motivated. It focused on the intense and fruitful interaction between learner and machine. It was still in use in the UK Post Office in the late 1960s.

 

Ecogame, developed in 1969, a management training game, had some success being bought by IBM. Targeted at business and political leadership, it was based on his ideas of self-organized, interactive learning in a simulated process and environment.

 

His teleprinter trainer caste CASTE, which stood for Course Assembly System and Tutorial Environment, was the size of a small room, built in 1972. It was an early adaptive learning system based on his learning theory. CASTE had an Entailment Structure or topic map to provide an overview, a Communications Module on a computer terminal, BOSS ( Belief and Opinion Sampling System) and a modelling system that would present learning and assessment tasks. Systems were designed for students with tutorial algorithms, to spot failure and reassess on lower levels of learning. It offered a range of choices to the learner. 

 

Thoughtsticker, built around the same time, used entailment meshes, was a hierarchical representation of knowledge. It originally had a set of pigeon holes with paper notes, later developed into an adaptive system that allowed users to move upwards on an entailment mesh of concepts. You could study it through a mesh of concepts with different perspectives. The learner would feed in concepts and the interconnections would be shown. A small computer was later used. It could also jump with was conditional hyperlinking. It was ahead of its time, unfortunately it was also mired in learning styles theory.

In all of this development and experimentation he was trying to implement his theory of learning - conversation theory. Teaching and learning were fundamentally conversations fo Pask, even when they involved technology.

 

Conversation theory

Pask studied learning through conversation, looking for identifiable features processes. Meaning is agreed use in the context of a conversation. This is not a fixed exchange of propositions but an exchange that shapes the thinking and learning of both parties. We may agree, disagree, modify beliefs, even change our minds but we gain from the conversation. Learners and teachers move through knowledge changing perspectives and levels. This, for Pask was the essence of learning. He sees conversation and learning as something active, in use, as we deal with the world. Conversation is also the link or knowledge we have with others, that they exist, like us. Conversation for Pask is at the root of everything we do, perception, emotional engagement and language. This is a much subtler idea than just verbal conversation. 

We learn together and are always using artefacts to learn and communicate. A conversation can be several different types, as an individual, two people, a group of people or combination of people and technology. Proximity is not the point, as technology destroys distance. This is a general conversational theory that sees people, their social interactions and technology as one.

Conversational theory saw learning as taking place in a learning environments, where we had sophisticated encounters with that environment, to it and back from it. We learn, Pask thought, by interacting, or having ‘conversations’, with people and our environment, including gestures, pictorial, with media and with and through machines. In that sense ‘conversation’ is almost metaphorical. We interact with the world in a way that is very much like ‘conversation’ in language.

At one level, teacher and learner can engage in a strict conversation, with interactions on two levels – how (teacher demonstrates) and learner does and why (explains) at the conceptual level. Pask saw that we engage in hierarchies of conversation but not in some simplistic Bloom hierarchy, we flit between levels. What he called styles of teaching or learning are like Wittgenstein’s language games, contextually different forms of conversation.

 

His learning theory saw conversation as the essential learning process. He studied conversations separately from the content of conversations to explore their processes and limits. Learning conversations are, and should be, open, complex and dynamic, not a clumsy, didactic movement toward a final truth. Variety, a plurality of ideas, perspectives and conversations matters, where learning emerges from the process of conversation.

 

Learning needs to provoked by a teacher and constructed by the learner and although Pask is rarely seen as a serious social constructivist, he did a lot of work, unlike most constructivists, on exactly how he though learners constructed knowledge and pragmatic skills. Importantly, he was sensitive to learning by doing and practical skills as well as knowledge.

 

In Conversation, cognition and learning (1975) he outlined the complexity of conversation, especially in learning, with hierarchical branching and loops, agreement, understanding, along with analogies and generalisations. Concepts are refined, expanded, altered, generalised, applied, related by analogy in the process of learning. This is not just dry rationalism, as for Pask to think is to feel, mind and body intertwined. He was keenly aware of our evolutionary legacy as biological beings. Conversation is how we think, and to teach and learn we must reproduce the process that evolution has bequeathed us. He had drawn from Vygotsky, the idea of a social being having developed throughout childhood, although Pask's focus was less on language, more on a wider social context.

 

Entailment meshes

Entailment meshes attempt to capture thought processes, with topics related to each other and grouped into relations, such as ‘analogies’ and ‘coherences’, that comprise concepts, which we share. We are the active constructers of concepts, making new finer distinctions, seeing similarities and differences, generalised, realised through action and interaction. All of this takes place naturally, we are conversational beings and have no choice in the matter. Conversation is an on-going, dynamic process, part of our being.

 

This is a radical departure from the database driven model that drove AI for so long, one of retrieval. We now interact with the world to learn through conversation. These conversations can be varied as they must suit the progress of the learner, not be pre-determined by too few constricting rules. He thought that most learning technology was hopelessly primitive in teaching and learning as they failed to recognise conversation as the means by which we learn. 

In learning there also has to be a contract to learn, common ground and action for successful learning, a process that is honest about contradictions, conflict, epiphanies and resolutions in the learning process. It is an agreed context.

Once a domain has been mapped out, one can also record the learner’s vectors through content, the routes taken, nodes reached. This is the sort of personalisation realised by modern adaptive systems and now generative AI, that can deliver the sort of sophisticated dialogue that Pask envisaged.

 

Interactions of actors

Pask and Gerard de Zeeuw took conversation theory one step further with the ‘Interactions of actors’ theory, where the scope was widened to include three or more actors in conversation, across time, entering and leaving such conversations. Here the defined process had knots, links, braids and a logic of the conversational and learning process but this is often metaphorical and lacks the complexity that AI models now have in capturing language. Rather then designing such a logic of process, AI learns from language and its use in Large Language Models, where the training data set is enormous. It is therefore empirically much more useful and representative of the reality of language use.

 

Generative AI

Conversations go on over time and become the culture, so LLMs capture many conversations. Conversations are captured as culture. So rather than constructing huge logic trees and cumbersome Paskian machines, Generative AI seems to deliver the richness of topics and language that allows it to deliver at least some of the conversational complexity that Pask thought was essential for learning. It already holds the relationships of words and phrases to each other by being trained in an unsupervised way, with a vast corpus of text, then further supervised training using RLHF (Reinforcement Learning from Human Feedback) and PPO (Proximal Policy Optimisation). Pask even talked about evil as the limits of the rights of actors to interact or hold conversation, prefiguring alignment and guard-railing.

 

The next stage is to provide some conversational structures to the dialogue that allow efficient learning to take place. This is precisely what is happening with Khanmigo, where structured feedback and encouragement adds motivational and useful feedback to the on-going learning conversation. Pask would have been delighted at the emergence of this type of learning system using Generative AI. 

 

Interaction is still primitive with computers, and we may have just seen a more conversational model emerge, where conversation, let’s call it on-going prompting, is replacing search and retrieval or request systems. Systems are becoming conversational, whether reading/writing or oral/aural and therefore more human. The technology must become more aware of people's needsto have more fruitful conversations. Conversations encapsulate interests, the seeking and learning process. They reveal intention.

 

The word ‘conversation’ may seem odd in the case of LLMs, as it lies beyond the idea of discrete individuals and stored knowledge into massive and sophisticated models with billions of parameters trained on unimaginably large sets of training data. But our interactions with these systems are certainly conversational, as learning is a process not an event. Pask realised that despite the radically different way these models work, they are still entities within a conversation with humans. The biological and technical are not separate but entwined in a single conversational system.

 

Conclusion

We would do well to pay more attention to learning through Pask’s conversational theory, with more empirical work on what works well in the context of increasing capability on the machine side through AI. Improvements that take the learning process into real conversations with structures that both allow for complexity but also guide towards outcomes, is now badly needed. This is what will give us useful, massively beneficial, teaching and learning. Let’s call them conversational systems, even a Universal Teacher.

Bibliography

Pask, G. (1961). [1968]. An approach to cybernetics. Hutchinson. 

Pask, G. (1962). a proposed evolutionary model. in H. von Foerster & G. W. J. Zopf (Eds.), Principles of self-organization: Transactions of the University of Illinois symposium on self-organization, Robert Allerton Park, 8 and 9 June 1960 (pp. 229–254). Pergamon Press. 

Pask, G. (1966). Men, machines and the control of learning. Educational Technology, 6(22), 1–12. 

Pask, G. (1968). a cybernetic model for some types of learning and mentation. in H. l. oestreicher & d. r. Moore (Eds.), Cybernetic problems in bionics (pp. 531–586). Gordon and Breach science Publishers. 

Pask, G. (1975). Conversation, cognition and learning: A cybernetic theory and methodology. Elsevier. 

Pask, G. (1975). The cybernetics of human learning and performance: A guide to theory and research. Hutchinson Educational. 

Pask, G. (1976). Conversation theory: Applications in education and epistemology. Elsevier.

Pask, G., & Curran, s. (1982). Microman: Living and growing with computers. Century Publising Co. 

Pask, G., & Kopstein, F. F. (1977). teaching machines revisited in the light of conversation theory. Educational Technology, 17(10), 38–41. 

Pask, G., & von Foerster, H. (1960). A predictive model for self-organizing systems.