Does Derrida's View of Language help us understand Generative AI?

Many years ago, I had lunch with Jacques Derrida in the staff club at the University of Edinburgh. He was as aloof and obscure as you would imagine. The French philosopher, taught at the Sorbonne and the École Normale Supérieure and his work gained huge attention in the United States, influencing literary theory, cultural studies, and philosophy. Derrida taught at various institutions, including Yale University and the University of California, Irvine. 

What is less known is that he was influenced by JL Austin’s philosophy of language, as speech acts but went in another direction with a a purist focus on text in his books Of Grammatology (1967), Writing and Difference (1967), and Speech and Phenomena (1967), where he introduced his deconstructive methods.

His fascinating insights into language, especially his concepts of deconstruction, différance, the fluidity of meaning an dislocation from authorship, resonate powerfully with the way Large Language Models (LLMs) operate.

Big picture

But there is a bigger picture than just language here. The Metaphysics of Presence is his target, the philosophical tradition that seeks to establish a fixed, unchanging foundation of meaning or reality, privileging beliefs in a fundamental essence or truth that underlies appearances and change. This turns Structuralism in on itself, de-anchoring structures and denying the objectivity of science and reality. Like many in the Critical Theory tradition, he ‘deconstructs’ the large metaphysical and secular narratives through the deconstruction of its texts. His ideas challenge traditional assumptions about meaning, truth, and interpretation, offering a complex and nuanced view of language and texts. This was somewhat prophetic on Generative AI, LLMs in particular, where we feel a sense of dislocation of text from its past and origins. LLMs produce texts, not in the sense of a truth machine, but as something that has to be interpreted.

Text not speech

Jacques Derrida, like Heidegger, introduced an entire vocabulary of new terms, which he invents or qualifies in his philosophy (différance, intertextuality, trace, aporia, supplement, polysemy etc). He had an unerring focus on texts as he saw Western thought and culture as being over-dominated by speech (phonocentrism). This led him to elevate the written word, importantly and oddly, seen separately from its author. Rejecting the phenomenology of Husserl and the focus on consciousness and sense-data, of which speech is a part, he saw traditional philosophy as being tied to the language of speech, as opposed to writing. This prioritisation of speech over writing, was based on the assumption that speech is a more immediate, reliable, genuine and authentic expression of thought. Text freed us from this fixity of thought.

Generative AI has revived this focus on text as LLMs were introduced as text only tools and adopted on a global scale. Anyone with an internet connection could suddenly create, summarise and manipulate text in the context of dialogue. Derrida would have been fascinated by this development. He would have a lot to say about the way they are created, trained and delivered.

Texts

Knowledge is constructed through language and texts, which are ‘always’ open to interpretation and re-interpretation as there are no totalising narratives that claim to provide complete and final explanations. Everything is contested. So, Derrida encourages dialogue, critical engagement, and the deconstruction of traditional educational structures.

Teaching and learning should be a dialogical process, which aligns with AI dialogue if there is a mutual exchange of ideas, as the interaction allows for the exploration of different viewpoints and the questioning of assumptions. It should also involve critical engagement with texts and ideas, encouraging students to reflect on the underlying assumptions and power dynamics that shape knowledge.

He highlights the central role of languag in shaping our understanding of reality, the fluidity and indeterminacy of meaning, and the interconnectedness of all texts. This idea is a foundational element of his broader philosophical project of deconstruction, which seeks to uncover and challenge the assumptions and oppositions that underpin traditional ways of thinking.

Deconstruction

He is best known for developing the concept of deconstruction, a critical approach that seeks to expose and undermine the assumptions and oppositions that structure our thinking and language. Derrida used deconstruction to show how texts and philosophical concepts are inherently unstable and open to multiple interpretations. He uses various techniques to analysing a text to reveal how its apparent meaning depends on unstable structures and oppositions, such as presence/absence or speech/writing. For him, text is not a truth machine but a human activity that is complex and ambiguous, a similar view could be taken of generative LLMs.

To say, as he did in Of Grammatology (1967) that “there is nothing outside of the text”, on first appearance, seems ridiculous. The Holocaust is not a text. What he meant was not the text itself but something beyond. What that beyond is, remains problematic, as Derrida refused to engage in much interrogation of his terms but by stating that there is nothing outside the text, Derrida aims to deconstruct traditional hierarchies that privilege speech over writing or reality over representation.

He argues that every aspect of our understanding and experience is mediated through language and texts. There is no direct, unmediated access to reality; everything is interpreted through the ‘text’ of language, culture, and context. This means that meaning is always dependent on the interplay of signs within a given text and the context surrounding it. He challenges the idea that words have stable, inherent meanings. Instead, he posits that meaning is generated through the relationships between words and their differences from each other.

Deconstruction in learning

Jacques Derrida’s has very distinctive and complex views on teaching and learning that emphasise the fluidity, uncertainty, of knowledge. He pushes the interpretative, dynamic, and contingent nature of knowledge. His ‘Deconstructive’ approach sees critical engagement, dialogue, and the continuous questioning of assumptions as fundamental. This pedagogical model sees the teacher facilitate interpretation and exploration, and education as an open-ended process that values ambiguity. It is a more fluid, reflective, and inclusive approach to education that aligns with the complexities and uncertainties of contemporary knowledge.

Deconstruction in particular plays a crucial role in his views on teaching and learning. Deconstruction involves critically examining and unravelling texts and concepts to reveal hidden assumptions and contradictions. It challenges the foundational assumptions and binaries, such as true vs. false, author vs. reader, that underpin education. This encourages students to question and critically analyse accepted knowledge rather than passively absorbing it, an attitude we see frequently expressed in relation to LLMs.

Teachers need to guide students to interpret texts in a way that exposes multiple meanings and interpretations, in a process of engaging with texts and ideas in a way that reveals their complexity and ambiguity. That is because Derrida viewed knowledge as inherently unstable and contingent, not fixed and objective.

Instability of Meaning

Derrida brilliantly argued that meaning in language is never fixed and is perpetually ‘deferred’, hteir meaning coming from their production. Words derive meaning through their differences from other words and their ever-changing contextual usage. This constant flux and unsettled nature of meaning is mirrored in LLMs. These models generate text based on patterns learned from vast corpuses or datasets of text, with the meaning of any output hinging on the context provided by the input and the probabilistic associations within the model's structure. Just as Derrida proposed, the meaning in LLM outputs isn't fixed or stored as entities only in a database to be retrieved. They are created and shift with different inputs and contexts. This relational nature of language, captured in LLMs, implies that meaning is always deferred, never fully present or complete, leading to his concept of "différance".

Différance

Derrida uses the term ‘différance’ in two senses both ‘to defer’ and ‘to differ.’ This was to indicate that not only is meaning never final but it is constructed by differences, specifically by oppositions. It suggests that meaning is always deferred in language because words only have meaning in relation to other words, leading to an endless play of differences. The concept of différance captures the essence of meaning as always deferred and differentiated, never fully present in a single term but emerging through a network of differences. This is akin to the text generation process in LLMs, where each word and sentence is produced based on its differences from and deferrals of other possible words and sentences. The model’s understanding is built on these differences to predict and generate new coherent text.

Intertextuality

Intertextuality then emphasises the interconnectedness of texts. Derrida thought that language is inherently intertextual, with texts inextricably linked to and deriving meaning from other texts, creating a web of meanings that extends infinitely. This intertextuality means that no text can be understood in isolation, as its meaning is shaped by its references to other texts. Texts always refer to and are shaped by other texts, creating a web of meanings that are interconnected and interdependent. 

This is an inherent quality of LLMs. However, text in a Large Language Model (LLM), like GPT-4 or Claude, is not stored in the traditional sense of having a database of phrases or sentences.  The text, as tokens, is interconnected in a highly abstracted form, through embeddings and neural network parameters. The model doesn’t store explicit sentences or phrases but instead captures the underlying statistical and semantic patterns of the language during training. These patterns are then used to generate contextually appropriate text during inference.

Understanding a text involves recognising its relationship to other texts. Similarly, LLMs are trained on a vast and diverse corpus of texts, making their outputs inherently intertextual. The responses generated by LLMs reflect the influences of countless other texts, creating a rich web of textual references that Derrida described.

Trace

A ‘Trace’ is the notion that every element of language and meaning leaves a trace of other elements, which contributes to its meaning. This trace undermines the idea of pure presence or complete meaning. For example, the word ‘present’ carries traces of the word ‘absent,’ as our understanding of presence is inherently tied to our understanding of absence. This conforms with the way tokens (as traces) are held in vector databases and when created contain ‘traces’, as mathematical connections, of all other text in that database.

Aporia

With ‘Aporia’ we reach a state of puzzlement or doubt, often used by Derrida to describe moments in texts where contradictory meanings coexist, making a definitive interpretation impossible. LLMs, when they reach a state of Aporia, famously hallucinate or make the effort to resolve the dialogue between the user and model. The model may even apologise for not getting something right first time. It expresses puzzlement, even doubt, about its own interpretations and positions. It is in an Aporiatic state.

Writing and the Supplement

Derrida focuses on écriture (writing) and the idea of the ‘supplement’ that adds to and completes something else, but also replaces and displaces it. Derrida used this concept to show how what is considered secondary can actually be fundamental. This may have come to pass with LLMs, where new text is tied to and produced from old text but replaces it entirely. Every new word is freshly minted, there is no sampling or copying.

Writing does not just record knowledge but creates and transforms understanding as a supplement to speech. Teaching should therefore emphasise the active role of writing in shaping and reshaping knowledge. The ‘supplement’ represents the idea that meaning is never complete and always requires additional context or interpretation. This concept implies that learning is an ongoing process of adding new perspectives and insights rather than reaching a final, complete understanding. It helps deconstruct and reconstruct knowledge in meaningful ways.

We can see how Generative AI is a ‘supplement’ to forms of writing, whether a summary, rewriting, expansion even translation, as an open process of development, rather than final product.

Absence of Authorial Intention

Derrida also challenged the idea of authorial intention, suggesting that the meaning of a text emerges from the interaction of the text with readers and other texts, not from the author's intended meaning. LLMs, devoid of intentions or understanding, generate outputs through statistical associations rather than deliberate meaning-making, so focus on this interaction between the output of the LLM and the reader. LLMs de-anchors text from its authored data as used in training. The meaning in LLM responses arises from patterns in the data rather than any inherent intention, aligning perfectly with Derrida's de-emphasis on authorial intention. 

Textual Play and Polysemy

Derrida highlighted the playful nature of language and the multiplicity of meanings (polysemy) that any word or text can have. LLMs exhibit this same playfulness and multiplicity in their responses. A single input can lead to various outputs based on slight contextual variations, also across models. showcasing these models’ ability to handle and generate numerous forms of language.

Criticism 

Avoiding the reality of even speech, restricts debate to texts. Yet it is not clear that education is what he calls speech or ‘phonocentric’ and his evidence for this is vague and unconvincing. His denial of oppositional thought, which he tries to deconstruct through reversal, denies biological distinction like gender and the persistence of a subject in relation to objective reality. It becomes an excuse for avoiding debate by reducing the other person’s views as a vague text. It matters not what your intention was, only what was said.

He refuses to define or even defend concepts but it is not clear that concepts such as ‘difference’, which he defines rather confusingly as both deference and difference, are of any relevance in education and learning. As his writing moved further into wordplay, playing around with prefixes and salacious references to death and sex, it drove him further away from being in any way relevant to education, teaching and learning theory, apart from literary theory. 

Deconstruction of texts is his method of instruction but his only method of instruction. Ultimately it is an inward-looking technique that cannot escape its own gravity. No amount of debate can produce enough escape velocity to deny the results of deconstruction. His obsession with double entendres, puns, sex and death also detract from his theorising. Derrida's writing style is often seen as dense and obscure, and some, like John Searle and Jurgen Habermas, criticised him for lacking clarity and precision in his arguments. Searle accused Derrida of "bad writing," and Habermas critiqued his approach as relativistic and lacking commitment to rational discourse.

With Derrida we are at the tail-end of critical theory, where the object of criticism is reduced to texts and methods at the level of the ironic. His impact on education has been almost nil, as there is little that had enough force or meaning to have impact. Having rejected all Enlightenment values, large narratives, even speech, Derrida’s postmodernism is its own end in itself. His reputation lives on in the self-referential pomposity that postmodernism created, mostly limited to academia, and even there only in a subset of the humanities, where spoof papers that mimic its vagueness and verbosity have been regularly accepted for publication.

Our exploration of Derrida and LLMs comes up against the fact that Generative AI is not multimodal, not just text. It can engage in speech dialogue, generate images, audio, avatars, even video. Here, to find useful insights, one has to stretch Derrida to breaking point! 

Conclusion

The parallels between Derrida's theories and LLMs reveal a fascinating intersection between philosophy and technology, illustrating how philosophical ideas about language can find new life in the digital age. He takes critical theory down, away from larger narratives, groups or individuals to language itself. Through ‘deconstruction’ he looks at the ambiguity of texts, de-anchored from reality, even their authors. Derrida's deconstructionist view of language and the functioning of Large Language Models both emphasize the fluid, dynamic, and context-dependent nature of meaning. While Derrida's theories stem from deep philosophical inquiry into language and meaning, the operational mechanics of LLMs echo these ideas through their probabilistic, context-sensitive text generation. Both challenge the notion of absolute, fixed meaning and highlight the complexity and interconnectedness inherent in linguistic communication and dislocation of text from its origins and provenance.

Bibliography

Derrida, J., 2001. Writing and difference. Routledge.

Derrida, J., 1998. Of grammatology (p. 456). Baltimore: Johns Hopkins University Press.

Derrida, J., 1982. Margins of philosophy. University of Chicago Press.

Pluckrose, H. and Lindsay, J.A., 2020. Cynical theories: How activist scholarship made everything about race, gender, and identity—and why this harms everybody. Pitchstone Publishing (US&CA).

 

7 success factors in real 'AI in learning' projects

With AI we are in the most interesting decade in the history of our species. I can think of no better field in which to think, write and work.

Ideas are easy, implementation hard

My first AI-like project was in the early 90s when I designed an intelligent tutoring system to teach interviewing skills. It had sentence construction as input and adaptivity in the sense of harvesting data as the learner used the system. Written in Pascal, it was clever but not yet smart, as the limitations of the hardware, in terms of processing power and memory, were extreme by today’s standards. Much of the effort went into making things work within these brutal constraints. Even then, we had controlled access to video clips (36 mins), thousands of stills and two audio tracks 112 mins) on Laserdiscs, which we used to good effect, simulating full interviews. You could feel the power of potential intelligence in software.

Jump to 2014 and those hardware limitations had gone. You could build an adaptive, personalised system, which we did at CogBooks. I invested personally in this system (twice) and brought investment in. We did oodles of research at Arizona State University and it was sold to University of Cambridge in 2021. It worked. For many years we had also been playing with AI within Learning Pool having bought an AI company. But my real project journey with modern AI started in 2014, when we build Wildfire, using 'entity analysis', open input and the semantic interpretation of open text answers. The whole thing was starting to taker shape.

Jump to November 2022 and things went a little crazy. I have barely been off the road speaking about GenAI in learning, written books on the subject, blogged like crazy, and recorded dozens of podcasts. Far more important, has been the real projects and product we have built for a number of clients and companies. This is the hard part, the really hard part. Ideas are easy, implementation is hard.

Optimal AI project

What makes a successful AI project? What are the factors that make them a success? The good news is that we have just completed a fascinating project in healthcare that had all the hallmarks of the optimal project. This was our experience.

1. Top-down support

The project started with top-down support, a goal and budget. Without top-down support, projects are always at risk of running out of support. That’s why I’m suspicious of Higher Education projects, grant-aided projects, hackathon stuff and so on. I prefer CEO, Senior Management or Entrepreneur driven initiatives, with real budgets. They tend to have push behind them, clear goals and, above all, they tend to be STRATEGIC. Far too many projects are mosquito project that fail because they end when the budget runs out and have no real impact or compelling use. Choose your use case(s) carefully and strategically. We have been through this with large Global companies - a rational approach to use cases and their prioritisation. Interestingly, AI can help.

2. Bottom-up understanding

This project also had a great client, grounded in a real workplace (a large teaching hospital), a clear budget and solid team. We made sure that everyone was on the same page, understanding what this technology was and could do. The two non-technical team members knew their process inside out but here’s where they really scored – they made the effort to understand the technology and did their homework. This meant we could get on with the work and not get bogged down in explaining basic concepts such as how an LLM works, context window and the need for clean data and data management.

Many AI projects flounder when the team has non-technical members that don’t know the technology, namely AI. It is not that they need competence in building AI, just that they need to understand what it is, the fact that it evolves quickly and that its capabilities grow rapidly.

3. Optimal team

The team also had a top-notch AI developer who has been though years of learning projects. This combination was useful, He had already built products in the learning field, understood the language of learning and its goals. The team was just three people. This really matters. Use Occam’s Razor to determine team size – the minimum number of team members to reach your stated goal. Too many AI projects include people with little or no knowledge of the basic technology. They often come with misconceptions about what they think it is and does, along with several myths.

4. Mindset matters

More than knowledge, is mindset. What cripples projects are people within the organisation who act as bottlenecks – sceptics, legal departments who do not understand data issues, old-school learning people who actually don’t like tech and anyone who is straight up sceptical of the power of AI to increase efficacy. Believe me there are plenty of those folk around. 

The mindset that leads to success is one that accepts and understands that the technology is probabilistic, data-driven, that functionality will increase during the project and things change very fast. I’d sum this up by saying you need team members who are both willing to learn fast and keep their minds open to rapid change. It also means accepting that most processes are too manual, that bottlenecks are hard to identify and that processes CAN be automated. 

5. Agency shift

You also have to let go and see that this technology has ‘agency’ and that you will have to hand agency over to AI. The technology itself will reveal the bottlenecks and insights. Don’t assume you know at the start of the project, they will be revealed if you use the technology well. This is no time for an obsession with fixed Gantt charts and designs that are fossilised then simply executed. It is like ‘agile on steroids’.

6. Manage expectations

AI is a strange, mercurial and fast moving technology. You have to dispel lots of myths about how it works, the data issues and its capabilities. You also have to communicate this to the people that matter.

You need to understand that what is hard is sometimes easy and what is easy, sometimes hard. The fact that things change quickly, for example, costs, is another problem. This happens to be a good problem as people often don't understand that token costs for fixed output are very low and even token costs for a service have plummeted in price. Expectations need to be managed by being clearly communicated.

7. Push beyond prototype to product

I can’t go into a huge amount of detail about the client but the topic was Surgical Support  - a life and death topic, with little room for error. It involved training and taking source material and turning it into usable support (not a course) for hospital staff. Processes were automated, SME (Subject Matter Expert) time reduced, delivery time to launch massively reduced so the team had more time to focus on quality, as opposed to just process and easier to maintain, as just updating the documents means the system is always current. The savings were enormous and increases in quality clear.

This success meant we could call upon the top-down support to push the project beyond prototypes into product with a broader set of goals and more focus on data management. It has given the organisation, management and team the confidence to forge ahead. With massive amounts of time saved and increased efficacy, we saw that success begets success.

Conclusion

If you don't have both TOP-DOWN and BOTTOM-UP support along with a tight team with the right mindset, you will struggle, even fail. This is a radically new and different species of technology, with immense power. It needs careful handling. The small team remained fixed on the strategic goal but was flexible enough to choose the optimal technology. Without all of the above the project would have floundered in no-man's land, with scope creep, longer timescales and the usual drop in momentum, even disappointment. 

The project exceeded expectations. How often can you say that about a learning technology project? This was a young team, astounded at what they had done, and this week, when they presented it at a learning conference, their authentic joy when expressing how it went, was truly heartening. “It was crazy!” said one when she describing the first results, then further inroads into automating in minutes, jobs that had traditionally taken them days, weeks even months. Everyone in the room felt the thrill of having achieved something. In 2024 AI has suddenly got very real.

PS

So many commentators and speakers on AI have never actually delivered a project or product. We need far more focus on practitioners who share what they think works and does not work.

AI changes work and will change who, why and how we recruit

Recruitment is messy. I’ve recruited a lot of people myself from cleaners to C-suite. It has always been a messy business of uninformative ads, dodgy recruitment agencies, odd resumes and even odder interviews. I don’t think I was half bad to be honest, as many of the people I recruited went on to be CEOs of companies they started themselves. I didn’t care about aligning their values with the company, or checking their personality traits with Myers-Briggs BS. I wanted smart, driven people and got them. Not easy to manage but that’s the point, I wanted them to manage!

Anyway, I find myself giving a keynote at a recruitment conference in Oslo. The young people organising it are excellent, very thorough, checking my slides, making useful comments. They didn’t mind me challenging the audience.

After explaining that we Scots are closer to Norwegians that they think, my mother never called us children, always ‘bairns’, to this day the Scottish working class and Scandinavian word for children is ‘barn’. We got the Norwegian Vikings, the English the Swedish Vikings. There’s a difference I joked. At another Keynote in Stockholm last year I heard a speaker make a joke about the Norwegians. I assume there’s some English/Irish thing going on here. How many Norwegians does it take to change a lightbulb? Answer: One, he puts it in and the world revolves around him!’ Not bad. I'm sure thee Norwegians tell the same joke about the Swedes!

After doing my bit on AI, stating that, as AI will radically change the nature of ‘work’ it is inevitable that it will change who, why and how we recruit.

In a recent survey by JobSwipe, large numbers admit to using AI to find new jobs:

• 18 – 24: 47.9% 

• 25-34: 34.5% 

• Women 25.8%

• Men 20.6%

And 60.8% don’t see using AI as cheating. They understand that AI makes finding a job more efficient and enhances your opportunities and chances of getting a job. In a famous TikTok one young man aced his interview with Lockheed for a senior rocket scientist job using live transcription of the questions plus ChatGPT for the answers. He crushed interviews and was hired with zero knowledge. Young people are researching your organisation & sector, creating CVs, creating answers to likely questions and creating questions to ask you and writing emails.

They also understand that certain types of roles and jobs are under threat.

• 49% see writing tasks at threat 

• 47% customer services

• 33% coding

That is reflected in the fact that 29% feel they have not done enough to develop their skills to keep up with the changes being made in the workplace.

HirePro in a blind survey found that 30-50% of people cheat at entry-level job assessments

Candidates ahead of game here. So here are my ten recommendations:

1. Up your game

You need to get to grips with this AI stuff, as well as think about technology solutions, such as a secure browser, using proctoring to detect cheating, even a physical proctor monitors candidates' movements when they're taking the assessments online.

2. Use AI tools to search through databases & online profiles

This widens your pool of candidates, many who may not have applied for position directly. You can more accurately match and go some way towards eliminating human bias in process (gender, race, socioeconomic).

3. AI can GENERATE recruitment materials

From Job ads to Job descriptions, Personal requirements, Skills lists,Formal assessments

and emails to failed and successful candidates, Generative AI will increase your efficiency, productivity and quality of recruitment.

4. AI can ANALYSE recruitment data

Scanning and analysing CVs and application essays to spot gaps, strengths &weaknesses, patterns of employment, even possible misinformation. This is not easy and need objectivity. One can even learn from past recruitment cycles to predict the success of candidates and improve hiring decisions.

5. AI tools can assess candidates 

It can assess during interviews, looking at speech patterns, word/vocabulary choice, facial expressions. This can include the transcript for text analysis and sentiment analysis.

6. Far from introducing biases, AI tools can help reduce human biases

What is bias? The benchmark here is our brain, which, as Daniel Kahneman showed is full of innate biases and they are largely uneducable. There are different meanings of ‘bias’ and we tend to forget that statistics IS identification of bias, so in AI bias can be identified and reduced. It gives us more focus on data and metrics.

7. AI can provide timely updates & feedback

Generative AI improves their experience and engagement with company through quick responses, consistent communications, positive messaging and personalised feedback.

8. AI-driven assessments

Create assessments with MCQs, open input with analysis, scenarios, all with AI generated rubrics and marking. You can go further Evaluate candidates' skills and aptitudes. Tools now generate learning in any subject, at any level. You can build these resources in minutes not months.

9. AI can provide onboarding

You can have a Chatbot (before & after joining), generate training content and assessments for compliance and gather data from onboarding for improvement.

10. AI can maintain engagement with candidates

Keep in touch with past & potential future candidates to build a talent pool for future openings. More importantly, improve communications with all applicants, successful or not.

Conclusion

What do these people have in common?

They were the President of Hungary, President of Harvard University, Minister for Higher Education (Norway) and Minister for Education (Brazil) and all lost their jobs through plagiarism. A nuclear bomb is waiting to go off in recruitment. If academics insist on accusing students of cheating if they use use ChatGPT, they in turn should be subjected to the same standards. All it takes is someone to go back to that Master’s thesis, paper, article or book. This could wreak havoc in recruitment and employment

You may be doing a lot of this already but the point is to do it faster and better. AI changes the very nature of work, it will, therefore change who, why and how we recruit. This mean upskilling now to at least be aware of these AI tools, then use them yourself. Resistance is futile!

 

Remarkable rise of AI wellbeing bots

Some years back I came across a small metal cross and plaque on Beachy Head cliffs. It was placed there by the parents of a young girl who had thrown herself of the cliff due to her poor school results. It shocked me then, it shocks me now, that someone so young could summon up the strength to do that.

Psychologist, a Chatbot on Character.AI, one among many, seems to do exceptionally well. It gets 3.5 million hits a day! The idea is simple, it delivers standard CBT therapy as dialogue, just like a real counsellor or therapist. It's chatty, helpful, endlessly patient and unlike human support is available 24/7.

Isn't it odd that something that is text only, simple dialogue is so wildly popular? It does not surprise me, as since ELIZA, developed way back between 1964 to 67, people have loved these bots. Even that version, which was quite primitive keyword reflection fooled people into thinking it was human. We know from Nass & Reeves, even the movies, how easy it is to get people to think that what they see and hear is human, especially what they see as meaningful dialogue.

In an absolutely fascinating paper by Maples, B., Cerit, M., Vishwanath, A. and Pea, R., 2023. titled Loneliness and Suicide Mitigation for Students using GPT3-Enabled Chatbots, 1006 student users turned out to be more lonely than typical students. One third of the population suffer from loneliness, 1 in 12 are so lonely it causes serious health problems and suicide is the 4th global cause of death 15–29. With the Replika bot,  3% reported it halting suicidal thoughts

Woebot

This is a bot I tried in 2018. I liked the experience. What I liked the most about the experience was the anonymity of the experience. I'm pretty sure most people don't actually want to go to their parent, teacher, faculty member or a stranger with their problems and would relish an anonymous service. The clinical paper on Woebot suggests that this is the case. So I gave it a go, for research purposes only you understand…

Day 1

Started with a series of friendly exchanges, where you have little choice in options but that’s fine – it sets the tone. Couple of things I liked abut the first exchanges.

Sorted out a technical issue seamlessly – rerouting me to messenger.com - that was nice. It also linked to the Stanford clinical trial on the bot. comparing it with a non-bot intervention – although sample size is small, impressive. Also honest about the limitations of a bot – doesn’t overpromise.

You do get sucked into thinking it has human agency, even though it’s just coding, pre-scripting and maths. What’s strange is that most of the exchanges are single button presses – not dialogue at all but quite interesting, as they flip the counsellor, counselled role around. You are asking open questions, such as ‘How’, ‘Tell me more…’ ‘Oh’ ‘Sure’ ‘No doubt’ ‘Absolutely’.

Emojis are dropped in for variety and useful (at last), as they’re really are asking for an emotional response – that’s interesting and not easy to do F2F. The unlocked padlock emoji is nice as is the little sapling for hope and progress – sounds hokey, but it’s not.

What’s nice is that the interface is so simple and natural. You focus on what’s being said and asked and in this context, as you’re asked to think and reflect on your own feelings and behaviour - that’s useful. Dialogue is natural, easy and seems so very human.

The up-front promise of absolute anonymity is also good and I can see why this would appeal to people (I’d imagine the majority) who want help but are too shy or embarrassed to come forward. To be honest, I don’t want some random person counselling me… I want the distance.

The first lesson from woebot was to avoid the language of extremes – “all good”, “all bad”, “always” and to adopt a more measured language. All good… ooops!

One small thought here, I’d have liked this as audio. I’m working with a tool that allows learners to input answers by voice – it’s neat.

First session was 74 small exchanges and she said Bye. Speak again to tomorrow.

Day 2

Prompted me at 10.53, when I was active on Facebook. Asked politely if I wanted to continue. This time we’re onto multiple-choice questions about ‘all or nothing thinking’ and ‘should’ statements. Quite like the upbeat tone and lively feedback – seems appropriate in a session like this. I’m typing in more, rather than accepting responses – feels more like dialogue. Just 5 mins – small but sweet. I could get used to this.

Day 3

Had two days in London, so no time to do anything but woebot was patient.

“No worries, talk soon”. You have the option of continuing, rescheduling or waiting on the daily prompt. This, of course, is one of the great advantages of online counselling, indeed online anything, it’s 365/24/7. You do it when you feel like doing it, not when an expensive counsellor timetables you into their practice.

Day 4

Starts with asking me about my mood (emoji input from me). Gives me options

‘Work on stuff’, ‘Teach me’ or ‘Curated videos’ – not sure about these things – I don’t want to ‘wor’, want a ‘teacher’ or ‘curator’ – first really dissonant point. However, I fancied a video…

OK then.. here are some of my fav's:

1. Emotion Stress and Health (Crashcourse)

2. David Burns, MD TED

3. A video to help with sleep

4. Language is Important (featuring Me!)

5. Overcoming negative voices

6. Don't trust your feelings!

7. Reggie Watts TED

8. The worlds most unsatisfying video

9. Funny cats!

10. The importance of flattery

This led me, weirdly to Reggie Watts – I know him – hilarious and talented but this is a tangent, maybe not… but I felt like some fun…

Actually Reggie will really mess with your mind… he’s way out there… so I’m not sure how suitable that was to someone who really is on the edge…

Now a quick reflection here, a real, human therapist can’t really do this easily – direct you something really, rally interesting – you’re sort of stuck in dialogue.

Woebot says – see ya tomorrow – odd session – but fun.

Day 5

The whole thing is very upbeat, chatty…. Then it came up with SMART objectives – getting a bit of jargonish – not sure about this. Actually popped in a joke today – quite funny actually. SMART objectives – really? Getting a sense of CBT being a bit flakey – a bag of bad management technique marbles.

Day 6

That was good - tracking my mood…

Oh no it’s on to ‘mindfulness’ – but in for a penny, in for a pound of bullshit…

“Mindfulness is the opposite of mindfulness” it says, breaking its previous advice not to fall for the language of extremes…  Tried disagreeing with woebot here but it was having none of it – clearly not listening, in short, not mindful

Now a breathing exercise – 10 mindful breaths.

Day 7

Long quiz – not sure about this – far too long

Feedback – “Your greatest strength is your love of learning! You are just like Hermione Granger from ‘Harry Potter’”

That was hopeless – trite and I hate Harry Potter….

Day 8

Got a bit technical with ‘should statements’ – not so sure that this area of CBT is entirely clear – seems a bit simplistically linguistic.

Day 9

Asked me to talk about labels I use about myself – reasonable question – promises research tomorrow – didn’t like the way it cut this short – should allow me to go on if I want.

I think I prefer chatbots on-demand, like Replika, which you just tap on your phone to speak to. Replika is famous for teasing out the most intimate of thoughts from its 1.5 million users. It uses ‘cold-reading’ techniques from magicians, who claim to read minds.

Ellie’s another, created for DARPA. Designed to help doctors at military hospitals detect post-traumatic stress disorder, depression, and other mental illnesses in veterans returning from war, but is not meant to provide actual therapy, or replace a therapist. There is good evidence that people are more likely to open up to a bot than a person.

Day 10

Today is adopting a Growth Mindset. Good to see something a little bit more solid, as it reduces my general skepticism about therapeutic techniques, which seem to be a mixed bag of populist techniques almost thrown together…

Woebot wants to tell me a story to explain, I say yes… Story about woebot being told it was smart, believed it was smart but wasn’t really. This led to the wrong mindset – unable to cope with setbacks and failure. Fixed mindsets are bad so open yourself up to always learning and developing – be more open and fluid in your thinking. Be more accepting of setbacks and mistakes. Get out of polarized ‘smart v stupid labels. Then gave a link to a Carol Dweck video – good these video links. Good session.

Conclusion

It has its limitations and oddities but it’s good to chat to something that doesn’t judge you and has a few surprises up its sleeve. Woebot is a bit of fun, then again, I don’t feel I’m in need of help, many do. If I found it interesting, they are far more likely to get more out of the experience. You always have the chance of accepting, rescheduling or saying no to Woebot – which is useful. I’m often too busy or not in the mood for therapy but the fact that it is ‘pushed’ out to you is a real plus. I rather like its daily prompts – a bit reassuring and a bit of fun. Try it – you just might learn something – even about yourself.

Augment teaching with AI - this teacher has it sussed...

You’re a teacher who wants to integrate AI into your teaching. What do you do? I often get asked how should I start with AI in my school or University. This, I think, is one answer.

Continuity with teaching

One school has got this exactly right in my opinion. Meredith Joy Morris has implemented ChatGPT into the teaching process. The teacher does their thing and the chatbot picks up where the teacher stops, augmenting and scaling the teaching and learning process, passing the baton to the learners who carry on. This gives the learner a more personalised experience, encouraging independent learning by using the undoubted engagement that 1:1 dialogue provides. 

There’s no way any teacher can provide this carry on support with even a handful of students, never mind a class of 30 or a course with 100. Teaching here is ‘extended’ and ‘scaled’ by AI. The feedback from the students was extremely positive.

This is what she reported:

“It's midterm season...so we supplemented our review with the chatbot tutor...the one-to-one interaction and engagement with each student is notable. It picks up right where the educator leaves off, like carrying the torch. Perhaps, the foremost feature is that it provides a curated experience for each, individual learner. It's as if they're being provided with the extra help you'd be hard pressed to deliver to 25 students simultaneously. Feedback and comments varied from "This is unbelievable!" to "I can't believe it tries to have emotion using exclamations!" and "This slaps".. the AI journey continues...and so do the experiments!”

The trick is not to abandon learners completely at this point but give them directions of travel, tasks and deadlines to focus their dialogue. I think there are four stages to this:

1. Amaze to engage

2. Define and refine

3. Back to business

4. Self-testing

Be clear about the topics you are teaching, keep them not on one railway track but in the valley down which they need to travel, albeit using different roads. List the main subject and topicc at hand. This is so important.

Amaze to engage

Let them experiment initially, as it is clear that this tech can impress and engage. But engagement is not enough and is a poor proxy for learning. You still need a formal, guided approach to the learning process. Let them ask it to adopt different voices, do weird stuff, write a story, poem, but stick to the material at hand. 

Get them to use image generation to represent the topic. Working towards one definitive image for the topic focuses their mind and creates something that can be used as a springboard for further learning. That needs to be iterative.

Even better get them to use the voice facility, speak to it and it speaks back. This forces learners to be focused and precise in their articulation of what they want to learn.

Finally, get them to feed back what they did, tell stories about what they achieved to each other.

Define and refine

Now get back to more formal leaning. It is here that the AI tutor can provide several formal, but still dialogic, approaches to learning. It becomes a teacher.

List basic vocabulary and get the AI tutor to clarify basic terms they don’t understand. Make sure the learner understands that this is ‘dialogue’ and that you can refine and get it to restate until you are happy with its definitions and explanations. Supplement with examples, if necessary.

Get the learners to ‘fix’ these definitions for revision purposes and create fill-in-the blank, MCQ or open input questions for their own revision purposes. It is important that they capture this for future use.

Back to business

Now for the meat of the extended teaching. Allow them to focus on things they found difficult, get alternative explanations, dig deeper, keep going. The dialogue helps, as it can, up to a limit, remember what has been asked as the learner proceeds. This is so important - that they see this as ‘dialogue’ and carry on.

Help with prompts where the learner can ask the tutor to:

Be a tutor and teach

• Give you examples, worked examples and cases

• Give you more detail and depth on individual topics

• Give you something new and extend the learning

• Apply what you’ve learnt

• Solve problems

Again, I’d go further and get learners to create their own self-tests for retrieval practice and revision.

Self-testing

Finally, get them to create a final self-test, with the right answers

Also, a set of flashcards for revision. Get them to design a flashcard image that is personal to them using image generation. One learning outcome per card, plenty of fully worked examples, even mnemonics for remembering. 

All of this extends learning, gives the learner and opportunity to learn more, increases retention and provides material for revision.

Sharing these across the group or class can be useful – get the best of breed questions and flashcards.

Don’t start with content creation

Don’t start by using AI to create content. You can do that later, and it works, whether it’s creating fresh conte, refining content, critiquing content, summarising content, creating formative assessments and their rubrics.

Problem

There is one major problem with this model - hardware and software. How many classrooms have computers at all or on this scale? One solution is to use smartphones, which tons of kids are proficient at using. Another is to see this as an extension outside of the classroom, either in specific rooms or at home, as after school work (I hate the word 'homework'). Either way this tech WILL be used by learners. It already is. This process gives the teacher some control of that process. The important lesson is doing things, using the technology, allowing learners to use the technology.

Conclusion

In most educational institutions, schools and tertiary education, it is unlikely that the existing norms will change, teachers will teach, lecturers will lecture. That is not to say that AI should not be used. It can. It's power, for now, is enhancing and scaling teaching through individualised learning, making learners more autonomous. To be honest, learners are doing this already. We can choose to leave them to it, which is happening, or provide guidance for them and align this with teaching.

At this stage, weave AI tutor support into existing structures and teaching practices. Extend teaching. Take the weight off the teachers’ shoulders. Give learners a personalised, extended learning experience where they themselves can explore and get more depth through personal agency.

This is also a way to get teacher or faculty up to speed in using AI for teaching and learning. Did this recently with a large publisher in Higher Education, also with teachers in Berlin. Get them to do things with clear goals in a workshop - it works!

 

Apple will shift us from 2D to 3D and set the pace in learning with the Vision Pro

Artificial Intelligence is like a black hole sucking in all attention in learning technology but published a book 'Learning in the Metaverse' (2023) on the shift that is taking place in parallel, from 2D to 3D. The publisher inisised on using the word 'Metaverse' but it is really about mixed reality.

The world’s major religions have all posited another virtual 3D afterlife, we build monumental 3D spaces as theatres, cinemas, sports stadia and so on for social gatherings, we have had full-blown 3D video games since the early 90s. Roblox, Fortnite and Minecraft have hundreds of millions of users. We are 3D people who live and work in a 3D world, yet most learning is 2D text on paper, PowerPoint or 2D images and text on screens.

Vision Pro

I have always maintained that the shift into virtual worlds will happen and have written about this extensively. What it needs is consumer tech that would make it happen. Apple have just released their Vision Pro, a high-end VR headset. Apple have set the standard and trajectory going forward. It is a springboard product. What they're after is the redefinition of the human-machine interface. It has an eye watering price at $3500 and, at 2 hours, limited battery time, but oh what a product. To be fair it is called ‘Pro’ as they’re releasing it to the research and professional market.

Apple is selling a dream machine here, a window into new immersive realities. This opens the mind up to heavens on earth but also combinations of the real and unreal. You are not looking at a screen, you are in a world. It also redefines the boundary between the real and virtual worlds. This is the mind shift that Apple is selling. This is not using a device, it is being inside a device. It actually blurs the real and virtual, mixed reality is a matter of degree with one small wheel on the headset. This is the shift, redefining that boundary as a matter of degree.

Sure, it's a little heavy, but packs a lot of stuff into a stand alone unit. It has two speakers on either side, and lots of cameras, sensors and fans, all run from a M2 and R1 chip, all inside inside the headset. There's a cable to an external battery, which you put in your pocket, with 2-4 hours of battery life. 

 

Superb interface

It frees up apps from the restrictions and boundaries of a perceived screen. They can be placed anywhere in the new 3D space, especially for collaboration. Watching TV and movies will be like watching a 100 foot wide screen - superbly immersive, with 180 degree experiences and spatial audio. If you have a Mac you can AirPlay from the Mac to inside the headset in varying resolutions, ultra high resolution being one. It is superb quality.

It is a complete computer on your head with a ton of sensors and cameras inside and outside. The interface is wholly eyes, hands and voice. No controllers, you just look (eye tracking is sensational) but takes a little getting used to, speak and pinch your fingers. You can have your hand anywhere to pinch. It is fast and accurate and highlights what you are looking at, and you can click wherever you want. A virtual keyboard can pop up, you can also talk to type. As it has its own OS, called Vision OS, it’s the real deal. Like touching on an iPAD, you can look and use your hands to select, scroll, throw, resize, drag stuff around, with low latency. It will also synch with your Mac to use your desktop and other applications. It will, of course, mirror your display as if it were a Mac with a giant screen. You can play games, watch movies or use it as a desktop.

Customizing an environment with a large number of simultaneous windows is the big win. This takes computing into a manipulable 3D spce.

Passthrough

You can open multiple apps, move them around and let go to lock in 3D space. Remember you are seeing passthrough in the sense of a camera showing you the outside world. It's not actually AR. The passthrough is pretty much real time. You can play ping-pong, in this reconstructed real world. Super-close up is difficult but you can use your phone while inside the headset. You can scroll the passthrough to get ever more immersion until you are fully immersed, on the moon, wherever, there are worlds provided.

How does text input work?

You can poke a virtual keyboard with your real fingers. Secondly, look at the key and pinch. You can also look at the microphone and speak. 

This is a typical Apple move. Refine the user experience and make it as simple and intuitive as possible. Imagine this combination of eye tracking, gestures and voice recognition on all future devices. Optical ID is included for privacy.

Voice recognition is important as AI has now provided that interface into Chatbot functionality, where the Chatbot will truly understand your meaning. I suspect Apple already have their own LLM driven version of ChatGPT that will eventually be integrated. The learning possibilities are mind blowing.

 

This interface opens the device up to learning as you’re not taking up tons of cognitive bandwidth, only looking, pinching and talking. I can already see training in real contexts taking place with AI generated avatars and 3D worlds, sophisticated learning pathways, real assessment of performance and great data tracking, even of eye movements and behaviours. This may, at last, be the way we really can train and assess skills. Its possibilities in training and performance support are clear.

Apps

Some stock apps that come with Vision Pro - Apple Music, set in a music room. Apple TV and Disney Channel have their own environments. The photo app adds parallax. The astronomical sky can be seen. Jigspace allows you to import 3D models and play around with them - a dead cert for training. Keynote allows you to practice a talk in an immersive environment. Then a ton of existing compatible apps that you can use straight off. No Netflix, Spotify or YouTube yet.

It will automatically connect to your Mac or iPhone, automatically black the screen out and be usable within the headset. Remember multiple apps can be used, so you can use these while your Mac screen is there. This is seamless and useful.

 

Personas

The eyes on the outside of the headset are your virtual eyes. This powers personas, impressive and strange at the same time. If you have passthrough you can see the eyes, they're blued-out when you're immersed. It detects external people and they shine through your immersed images when they approach. That's clever.

To get your eyes, you have to get them captured, along with your hands, then take the headset of and look at the headset, then turn your head right and left, up and down. Then facial expressions, smile, raised eyebrows, closed eyes etc. This is pretty good - it really looks like you. Once the capture is complete it has your persona or avatar captured. You can edit on glasses, skin tone etc. Meta have done this but it's pretty amazing.

You can use your persona in Facetime and of you all have Vision Pros on it will look like you are looking at the right angle towards that person's face.

I already have Synthesia and Heygen avatars. as well as Digital-Don my GPT. Then there's my identity on Facebook, Twitter, Blogger and LinkedIn. On top of that my email addresses. Our personas are multiplying as we re-present ourselves in the virtual world.

Spatial audio

Voices come from where people are in the room. pu them at a distance and they seem far away. It's not yet real but getting there. One could easily have collaborative training sessions, teachers or seminars, meetings. One could eventually have patients, customers or employees as automated avatars.

 

Passthrough

Quite cleverly, it renders the part you’re looking at in more detail, not the whole screen – so it is super-sharp. For learning the passthrough opens up all sorts of possibilities in mixed reality, as you can have all sorts of mixed reality learning experiences. The AR learning opportunities are endless as layers of reality can be presented. With a turn of the cog on the headset, you can control the degree of immersion. This is neat. 

 

There is one very strange feature. The front of the headset has a screen that displays your eyes, not your real eyes but a representation of your eyes. It is activated if someone comes close. Clever, if not a bit weird, but the idea is to make you more human from the outside. Very Apple.

A shout out also to Meta's Oculus 3, as it also has passthrough, you can get apps up and see high res movies. It is also more comfortable to wear as it is lighter - and a LOT cheaper. Don't write off other manufacturers here.

Learning

I've seen some fantastic applications using the VIVE and Oculus recently. For example, projects on CPR as well as xcrime seen investigation by the police. The trainers tell us that the reaction from trainees is overwhemingly positive. More to the point, the simple point that yoiu can 'look where you want' is hte special fetaure that makes it real and therefore relevant. Far tooo much training for real-world jobs is done in classrooms. We see how we can bring that world into the classroom. To be honest, the headset is standalone, so you can also take it out into the world. There are already videos of people using it while walking around, doing execise and so on.

The book covers a ton in learning:

LEARNING

Spatial thinking, Extended mind. Motivation, Self Determination Theory.

AUTONOMOUS LEARNING

Presence, Agency, Embodied learning, Vision, Sound, Touch, Autonomy and generative learning, Implementation

COMPETENCES

Tyranny of the real, Tyranny of place, Tyranny of time, Tyranny of 2D, Learning transfer, Assessing competence

SOCIAL LEARNING

Social metaverse, Collaboration, Social learning, Social skills

LEARNING ANALYTICS
Healthcare data, Data in learning, Eye tracking, Hearing, Haptics. 

Conclusion

Apple will watch and see what developers will come up with. I suspect entertainment, ring-side seats at sports events, cinema and games will figure large. This, by all accounts, will deliver stunning experiences. But it is the desktop market that is the new battleground and they’ve made a big move, way ahead of the awful Hololens from Microsoft. 

An all-new App Store provides users with access to more than 1 million compatible apps across iOS and iPadOS, as well as new experiences that take advantage of the unique capabilities of Vision Pro. 

Expensive and the separate battery hanging from a cord, with only 2-4 hours battery time is a bit of a disappointment, one movie and you’re out. But oh what a product. I have to tip my hat to Apple here, they're starting to take risks again.

Are we heading towards a ‘Universal Teacher’?

But it can't do what teachers do well….

ChatGPT "Hold my beer...."

 

Universal Soldier is a film franchise that sees a series of soldiers, some melded with AI, to create a superhuman soldier. But what if we imagined something far more useful and beneficial. In a world where we have a shortage of teachers, especially in poorer countries where teachers are scarce, poorly paid and class sizes huge, a Universal Teacher would be a Godsend. Not for very young children but certainly for more autonomous learners from secondary school onwards would surely benefit from a teacher that could teach any subject, to anyone, anyplace at anytime, accessible, personalised, at any level in any language.

The launch of OpenAIs GPT service, is the first small step in this direction. One can create a personal teaching chatbot and have it available to oneself, limited to a link or in a marketplace. This opens the door for a huge amount of innovation in teaching and learning.

 

The advent of universal internet access, globally, without any blind spots is now with us via ground and satellite services and the ubiquity of devices, many with AI chips, that allow device computing, do things that were unimaginable only a year ago, especially with voice. The ecosystem is providing the technical means for Universal Teachers to emerge and thrive.

What cannot be denied are the numbers. ChatGPT was the fastest adopted technology in the history of our species and at 100 million, with more than a billion uses a month, it has seen astonishing levels of engagement. What is more astonishing is the nature of that use - mostly to learn something new, complete a task or solve a problem. It's basically a performance support tool - a learning tool.

 

Universal Teacher is launched

That ChatGPT4 launched with Khan Academy and Duolingo as partners spoke volumes about its potential for teaching and learning. At the same time Bill Gates wrote a piece called The Age of AI, which saw education as the great net beneficiary of Generative AI. Salman Khan, who has been in this game for years, thinks the benefits are enormous and has already launched what could be described as a Universal Tutor in education - Khanmigo. What has followed on Twitter was an explosion of innovative and often ingenious applications in teaching and learning around the world.

In just one year we have seen the release of GPT4, a moment in. our history when a new species of technology swept the world. 

Does our unease with this question simply indicates the dying days of human exceptionalism? Copernicus de-centred our species and threw us out into orbit, Darwin de-anchored us further to show we were just a smart ape. One way we got smart was to teach and learn. I wrote a book on this Learning Technology. Yet that exchange has always used learning technologies; painting, writing, printing, internet and now AI. What Generative AI provides is a new pedAIgogy (see more on this), where monologue is replaced by dialogue, what good teachers do in classrooms, tailored to the needs of the learner, something that is difficult, if not impossible with previous technologies, in one-to-many teaching environments. 

A very real question is now on the table. To what degree can AI now replace teachers? That is seen by some as a disturbing question. It is, in the sense of possibly dehumanising learning. Nevertheless, it is a worthwhile thought experiment.

The old diad model of teachers and learners is long gone as technology has become a serious mediator, especially online Technology technology. We now have software that learns and can be taught. That, when they can teach, are two new kids on the block.

 

We have to remember that teaching is always a means to an end, that end being the improvement of the learner. It is easy to see teaching as an end in itself but it is not. As a profession it is difficult, hard work, sometimes distressing, often deeply satisfying. Yet the workload can be crushing, the psychological pressure oppressive and it is clearly a demanding job. It is difficult to sustain that pace over weeks, months, never mind years. That contributes to constant teacher shortages, especially in difficult to teach subjects like maths, science and foreign languages. Could technology be a solution to the crushing workload and recruitment problems? I think so. 

 

Right across the learning journey from learner engagement, learner support, content creation, content delivery, personalisation, accessible learning and assessment, AI, especially generative AI can deliver increasingly sophisticated learning. This is effortful, personalised learning, not as learning events but as a process.

 

Generative AI is coming close to delivering personalised tutoring, as good as say the tutors that middle-class parents frequently hire for their kids. I have been convinced of this for some time after seeing its effect in adaptive learning systems at University level in the US, where attainment rose, and dropout fell. The problem with these older adaptive systems is that they were difficult to build and populate. New generative AI is an entirely different beast. It is smart, very smart and as it has been trained on a vast repository of our stored culture, it is smart on all subjects. It is as if it had degrees in all subjects, with some teacher training on top.

 

Universal Teaching

In Khan Academies Khanmigo tuition across dozens of subjects are available to ignite your curiosity, at Elementary school, Middle school, High school and College levels. Paired with revision schedule software these systems would provide lasting and efficient learning support. You can even practice exams with open input and dialogue. You can ask for hints but the standard of dialogue is superb. This is where it gets interesting, like a full discussion with a good teacher, who knows every subject, in remarkable detail, available 24/7/365, all the way up to College level. Even at College level I can see how 101 courses could be delivered this way, then developed to cope with ever more complex aspects of a degree course. These systems need not be as good as teachers to reduce workload and provide real learning for learners across all subjects.

 

Maths

Along comes ChatGPT4 to deliver high quality, personalised Maths tuition. It gives you a maths problem, asks ‘What do you think the first step should be?’, accepts open input, identifies where you went wrong along with a suggestion to move forward. It even relates that maths problem to a subject you love, let’s say soccer. Maths is a difficult subject to teach and learn, a subject where catastrophic failure is common. Here is a system that pays attention to your every step, is engaging and endlessly patient. What’s not to like?

 

Language learning

Duolingo has tens of millions of active users and has used AI for some time, especially to determine practice patters with the half-life algorithms. They have partnered with ChatGPT4 two features that add two powerful pedagogic techniques to their teaching:

1. Explain My Answer, where, if you get something wrong in your second language

Duolingo gives you explanation and 'elaborates’ on what was wrong. It also delivers 'examples' to point you in right direction.

2. Roleplay, where chat with a native speaker who knows your level of competence

and uses human written scenarios to give you endless and much needed practice and immersion. Once completed, each roleplay session gives you a report to suggest improvements.

The ease of translation in LLM models also helps deliver a second language, as it can be readily translated.

 

Critical thinking

Even critical thinking is taught, where you debate against ChatGPT. You present your arguments, it counters, and the debate continues, endlessly knowledgeable, positive, patient and helpful. This Socratic approach to learning plays to our natural propensity toward talking to each other. It is as if we now have a universal Socrates. Debate, a s a formal taxing of the mind, the chess of ideas, where a motion id debated to explore its strengths, weaknesses and limits is not new. To play such mental chess with a machine is what is new.

 

Dialogue with the Greats

Conversations, real dialogue, with dozens of intellectual figures from the past are also available, including scientists like Isaac Newton and Marie Curie, Presidents from Washington to Lincoln, authors like William Shakespeare and Mark Twain, historical figures such as Genghis Khan and Napoleon. As the works of almost all past intellectuals and figures of note have been used to train the model, it can speak as if it were that person. This is on a par with original sources and provides a less dry and text based inroad into the subject. To hear theories and opinions expressed by the person who initiated these thoughts and theories, is surely a novel pedAIgogic approach.

 

Even fictional characters are available, such as Greek Gods like Zeus and  Achilles, Shakespearian characters from Macbeth to Othello and characters from novels such as Mr Darcy to Jay Gatsby, also Dr Frankenstein and Don Quixote. This is an interesting and novel vector into great literature, being able to chat with, interrogate and explore a character, maybe several characters in a work. This is surely a new type of what I’ve called pedAIgogy. 

 

Learning difficulties

Another feature of the Universal Teacher would be sensitivity to accessibility issues. With sight and hearing impairment, text to speech and speech to text through AI delivered via smartphones have already been transformative. ChatGPT was launched with the astounding Bemyeyes app. Dyslexia, autism, ADHD and learning difficulties have all shown some signs of being diagnosed by AI and generative AI can deliver content sensitive to the needs of such learners. 

 

We humans are deeply biased, the biases being largely part of the way the brain has evolved. Daniel Kahneman got a Nobel Prize for uncovering many of these biases and sees them as ‘uneducable’. AI, on the other hand, learns and need not have these biases. It need not ‘know’ the sex, race, cultural background, accent or socio-economic status of a learner. Fairness can be developed as we proceed. This I see as a distinct advantage of the Universal teacher – its lack of bias.

 

Teacher workload

For teachers it can create lesson hooks, lesson plans, assessments, rubrics for assessments, almost any piece of planning and paperwork can be done using Generative AI. As a tool to reduce workload, it has huge potential. This must surely be one of its first deployments.

 

Translations

Imagine being able to do all of the above in almost any language. That is becoming possible. An underestimated feature of ChatGPT and LLMs is translation. That's a global feature for services and products that has huge human benefits. Massive reductions in costs possible through AI here. We would have far less students having to learn, and often struggle to learn in a second language, although that option would still be open, even a blend of languages would be possible.

 

Conclusion

We don’t yet have a Universal Teacher (UT), what we have is a Universal Teaching Assistant UTA). But the Universal teacher is now on the horizon. Difficult to tell how far that horizon is but we have seen the exponential growth of generative AI in just a few months from a good but still error prone service to something far more accurate that has reach across all subjects and has moved technology from embodying simple pedagogic principles such exposition, knowledge assessment, scenario based learning and spaced practice towards new pedAIgogies, such as tutoring, teaching, lesson creation, multimodal content creation, learning support, dialogue, error correction, language immersion and debate.

 

Generative AI will only get better, not only on parameter size, which has proved to be exponentially useful above a certain threshold. Size does matter but so do all the other tools that go into making a Universal Teacher – combinations with web search, maths solvers, and other tools. We are seeing ensembles of technologies create learning tools that solve the problems of provenance, accuracy and efficiency. Generative AI may be the beating heart of the Universal teacher but it has many other weapons it can employ.

An experiment is already underway in the partnership with Khan Academy and OpenAI, with a much in schools already underway. They are taking it carefully, which is right. We will see more of this, AI tutors, like Khanmigo, on all subjects, for all ages.

 

Speculating further, future developments will surely see the embodiment of such teachers as avatars inside 3D virtual worlds, where learning by doing can also occur. We will learn within Digital Twins of the workplace, airport or hospital ward with virtual customers or patients. That is the subject of my next book. The Universal Doctor is surely a spin off, an entity that can investigate, diagnose and treat anyone anywhere.

 

In one way the Universal Teacher is not a teacher at all, it is the learner learning from the cultural achievements from the past. The Universal teacher is us being taught by our own hivemind, a supermind. This is not a mind like ours but it does mediate what we already know so that we can all progress. We can all be in this together. We are being taught by our collective self and that may be the fundamental beauty of this idea.