Saturday, December 23, 2017

Is debate around 'bias in AI' driven by human bias? Discuss

When AI is mentioned it is only a matter of time before the word ‘bias’ is heard. They seem to go together like ping and pong, especially in debates around AI in education. Yet the discussions are often merely examples of bias themselves – confirmation, negativity and availability baises. There’s little analysis behind the claims. ‘AI programmers are largely white males so all algorithms are biased - patriarchal and racist’ or the commonly uttered phrase ‘All algorithms are biased’. In practice, you hear the same few examples being brought up time and time again: black face/gorilla, recruitment and reoffender software. Most of these examples have their origin in Cathy O’Neil’s Weapons of Math destruction orminternet memes. More of this later.

To be fair AI is for most an invisible force, the part of the iceberg that lies below the surface. AI is many things, can be opaque technically and true causality difficult to trace. So, to unpack this issue it may be wise to look at the premises of the argument, as this is where many of the misconceptions arise.

Coders and AI
First up, the charge that the root cause is male, white coders. AI programmers these days are more likely to be Chinese or Indian than white. AI is a global phenomenon, not confined to the western world. 

The Chinese government has invested a great deal in these skills through Artificial Intelligence 2.0. The 13th Five-Year Plan (2016-2020), the Made in China 2025 program, Robotics Industry Development Plan and Three-Year Guidance for Internet Plus Artificial Intelligence Plan (2016-2018) are all contributing to boosting AI skills, research and development. India has an education system that sees ‘engineering’ and ‘programming’ as admirable careers and a huge outsourcing software industry with a $150 billion IT export business.
Even in Silicon Valley the presence of Asian and Indian programmers is so prevalent that they feature in every sitcom on the subject. Even if the numbers are wrong the idea that coders infect AI with racist code, like the spread of Ebola, is ridiculous. One wouldn’t deny the probable presence of some bias but the idea that it is omnipresent is ridiculous.

Gender and AI
True there is a gender differential, and this will continue, as there are gender differences when it comes to focused, attention to detail coding in the higher echelons of AI programming. We know that there is a genetic cause of autism, a constellation (not spectrum), of cognitive traits and that this is weighted towards males (and no it is not merely a function of underdiagnosis in girls). For this reason alone there is likely to be a gender difference in high-performance coding teams for the foreseeable future. In addition, the idea that these coders are unconsciously, or worse, consciously creating racist and sexist algorithms is an exaggeration. One has to work quite hard to do this and to suggest that ALL algorithms are written in this way is another exaggeration. Some may, but most are not.

Anthropomorphic bias and AI
The term Artificial Intelligence can in itself be a problem, as the word ‘intelligence’ is a genuinely misleading, anthropomorphic term. AI is not cognitive in any meaningful sense, not conscious and not intelligent other than in the sense that it can perform some very specific tasks well. It may win at chess and GO but it doesn’t know that it even playing these things, never mind the fact that it has won.

Anthropomorphic bias appears to arise from our natural ability to read the minds of others and therefore attribute qualities to computers and software that are not actually there. Behind this basic confusion is the idea that AI is one thing – it is not – it encapsulates 2500 years of mathematics since Euclid put the first algorithm down on papyrus and there are many schools of AI that take radically different approaches. The field is an array of different techniques, often mathematically, quite separate from each other.

ALL humans are biased
First, it is true that ALL humans are biased, as shown by Nobel Prize winning psychologist Daniel Kahneman and his colleague Amos Tversky, who exposed a whole pantheon of biases that we are largely born with and are difficult to shift, even through education and training. Teaching is soaked in bias. There is socio-economic bias in policy as it is often made by those who favour a certain type of education. Education can be bought privately introducing inequalities. Gender, race and socio-economic bias is often found in the act of teaching itself. We know that gender bias is present in subtly directing girls away from STEM subjects and we know that children from lower socio-economic groups are treated differently. Even, so-called objective assessment is biased, often influenced by all sorts of cognitive factors – content bias, context bias, marking bias and so on.

Bias in thinking about AI
There are several human biases behind our thinking about AI.

We have already mentioned Anthropomorphic bias, where reading ‘bias’ into software is often the result of this over-anthropomorphising.

Availability bias arises when we frame thoughts on what is available, rather than pure reason. So crude images of robots enter the mind as characterising AI, as opposed to software or mathematics, which is not, for most, easy to call to mind or visualise. This skews our view of what AI is and its dangers, often producing dystopian ‘Hollywood’ perspectives, rather than objective judgement.

Then there’s Negativity bias, where the negative has more impact than the positive, so the Rise of the Robots and other dystopian visions come to mind more readily than positive examples such as fraud detection or cancer diagnosis.

Most of all we have Confirmation bias, that leaps into action whenever we hear of something that seems like a threat and we want to confirm our view of it as ethically wrong.

Indeed, the accusation that all algorithms are biased is often (not always) a combination of ignorance about what algorithms are and a combination of four human biases – anthropomorphism, availability, negativity, confirmation and anthropomorphism bias. It is often a sign of bias in the objector, who wants to confirm their own deficit-based weltanschauung and apply a universal, dystopian interpretation to AI with a healthy dose of neophobia (fear of the new).

ALL AI is not biased
You are likely in your first lesson on algorithms to be taught some sorting mechanisms (there are many). Now it is difficult to see how sorting a set of random numbers into ascending order can be either sexist or racist. The point is that most algorithms are benign, doing a mechanical job and free from bias. They can improve performance in terms of strength, precision and performance over time (robots in factories), compressing and decompressing comms, encryption algorithms, computational strategies in games (chess, GO, Poker and so on), diagnosis-investigation-treatment in healthcare and reduced fraud in finance. Most algorithms, embedded in most contexts are benign and free from bias.

Note that I said ‘most’ not ‘all’. It is not true to say that all algorithms and/or data sets are biased, unless one resorts to the idea that everything is socially constructed and therefore subject to bias. As Popper showed, this is an all-embracing theory to which there is no possible objection, as even the objections are interpreted as being part of the problem. This is, in effect, a sociological dead-end.

Bias in statistics and maths
Al is is not conscious or aware of its purpose. It is, as Roger Schank kept saying, just software, and as such, is not ‘biased’ in the way we attribute that word to ‘humans’. The biases in humans have evolved over millions of years with additional cultural input. AI is maths and we must be careful about anthropomorphising the problem. There is a definition of ‘bias’ in statistics, which is not a pejorative term, but precisely defined as the difference between an expected value and the true value of a parameter. If the value is zero, it is called unbiased. This is not so much bias as a precise recognition of differentials.
However, human bias can be translated into other forms of statistical or mathematical bias. One must now distinguish between algorithms and data. There is no exact mathematical definition of ‘algorithm’ where bias is most likely to be introduced through weightings and techniques used. Data is where most of the problems arise. One example is poor sampling; too small a sample, under-representations or over-representations. Data collection can also have bias due to faulty data gathering in the instruments themselves. Selection bias in data occurs when it is gathered selectively and not randomly.

However, the statistical approach at least recognises these biases and adopts scientific and mathematical methods to try to eliminate these biases. This is a key point – human bias often goes unchecked, statistical and mathematical bias is subjected to rigorous checks. That is not to say that it is flawless but error rates and attempts to quantify statistical and mathematical bias have been developed over a long time, to counter human bias. That is the essence of the scientific method.

An aside…
The word ‘algorithm’ induces a sort of simplistic interpretation of AI. Some algorithms are not created by humans, code can create code, some are deliberately generated in evolutionary AI to create variation and then selection against a fitness purpose. It’s complex. There are algorithms in nature that determine genetic outcomes, the way plants grow and many other natural phenomena. Some thing that there is a set of deep algorithms that determine the whole of life itself. Evolutionary AI allows algorithms to be promulgated or generated by algorithms themselves, in an attempt to mimic evolution, but defining fitness and selecting those that work. While it is true that bias can creep into this process it is wrong to claim that all algorithms are created solely by the hand of the coder.

AI and transparency
A common observation in contemporary AI is that its inner workings are opaque, especially machine learning using neural networks. But compare this to another social good – medicine. We know it works but we don’t know how. As Jon Clardy, a professor of biological chemistry and molecular pharmacology at Harvard Medical School says, "the idea that drugs are the result of a clean, logical search for molecules that work is a ‘fairytale'”. Many drugs work but we have no idea why they work. Medicine tends to throw possible solutions at problems, then observe if it works or not. Now most AI is not like this but some is. We need to be careful about bias but in many cases, especially in education, we are more interested in outputs and attainment, which can be measured in relation to social equality and equality of opportunity. We have a far greater chance of tackling these problems using AI than by sticking to good, old-fashioned bias in human teaching.

Fail means First Attempt In Learning
Nass and Reeves through 35 studies in The Media Equation showed that the temptation to anthropomorphise technology is always there. We must resist the temptation to think this is anything but bias. When an algorithm, for example, correlates a black face with a gorilla, it is not that it is biased in the human sense of being a racist, namely a racist agent. The AI knows nothing of itself, it is just software. Indeed, it is merely an attempt to execute code and this sort of error is often how machine learning actually learns. Indeed, this repeated attempt at statistical optimisation lies at the very heart of what AI is. Failure is what makes it tick. The good news is that repeated failure results in improvement in machine learning, reinforcement learning, adversarial techniques and so on. It is often absolutely necessary to learn from mistakes to make progress. We need to applaud failure, not jump on the bias bandwagon.

When Google was found to stick the label of gorilla on black faces in 2015, there is no doubt that it was racist in the sense of causing offence. Rather then someone being racist in Google, or having a piece of maths that is racist in any intentional sense, this is a systems failure. The problem was spotted and Google responded within the hour. We need to recognise that technology is rarely foolproof, neither are humans. Failures will occur. Machines do not have the cognitive checks and balances that humans have on such cultural issues but they can be changed and improved to avoid them. We need to see this as a process and not just block progress on the back of outliers. We need to accept that these are mistakes and learn from these mistakes. If mistakes are made, call them out, eliminate the errors and move on. FAIL in this case means First Attempt In Learning. The correct response is not to define and dismiss AI because of these failures but see them as opportunities for success.

The main problem here, is not the very real issue of emanating bias from software, which is what we must strive to do but the simple contrarianism behind much of the debate. This was largely fuelled by one book….

Weapons of 'Math' Destruction - sexed up dossier on AI?
Unfortunate title, as O’Neil’s supposed WMDs are as bad as Saddam Hussein’s mythical WMDs, the evidence similarly weak, sexed up and cherry picked. This is the go-to book for those who want to stick it to AI by reading a pot-boiler. But rather than taking an honest look at the subject, O’Neil takes the ‘Weapons of Math Destruction’ line far too literally, and unwittingly re-uses a term that has come to mean exaggeration and untruths. The book has some good case studies and passages but the search for truth is lost as she tries too hard to be a clickbait contrarian.

Bad examples
The first example borders on the bizarre. It concerns a teacher who is supposedly sacked because an algorithm said she should be sacked. Yet the true cause, as revealed by O’Neil, are other teachers who have cheated on behalf of their students in tests. Interestingly, they were caught through statistical checking, as too many erasures were found on the test sheets. That’s more man than machine.

The second is even worse. Nobody really thinks that US College Rankings are algorithmic in any serious sense. The ranking models are quite simply statistically wrong. The problem is not the existence of fictional WMDs but poor schoolboy errors in the basic maths. It is a straw man, as they use subjective surveys and proxies and everybody knows they are gamed. Malcolm Gladwell did a much better job in exposing them as self-fulfilling exercises in marketing. In fact. most of the problems uncovered in the book, if one does a deeper analysis, are human.

Take PredPol, the predictive policing software. Sure it has its glitches but the advantages vastly outweigh the disadvantages and the system, and its use, evolve over time to eliminate the problems. The main problem here is a form of bias or one-sidedness in the analysis. Most technology has a downside. We drive cars, despite the fact that well over a million people die gruesome and painful deaths every year from in car accidents. Rather than tease out the complexity, even comparing upsides with downsides, we are given over-simplifications. The proposition that all algorithms are biased is as foolish as the idea that all algorithms are free from bias. This is a complex area that needs careful thought and the real truth lies, as usual, somewhere in-between. Technology often has this cost-benefit feature. To focus on just one side is quite simply a mathematical distortion.

The chapter headings are also a dead giveaway - Bomb Parts, Shell Shocked, Arms Race, Civilian Casualties, Ineligible to serve, Sweating Bullets, Collateral Damage, No Safe Zone, The Targeted Civilian and Propaganda Machine. This is not 9/11 and the language of WMDs is hyperbolic - verging on propaganda itself.

At times O’Neil makes good points on ‘data' – small data sets, subjective survey data and proxies – but this is nothing new and features in any 101 statistics course. The mistake is to pin the bad data problem on algorithms and AI – that’s often a misattribution. Time and time again we get straw men in online advertising, personality tests, credit scoring, recruitment, insurance, social media. Sure problems exist but posing marginal errors as a global threat is a tactic that may sell books but is hardly objective. In this sense, O'Neil plays the very game she professes to despise - bias and exaggeration.

The final chapter is where it all goes badly wrong, with the laughable Hippocratic Oath. Here’s the first line in her imagined oath “I will remember that I didn’t make the world, and it doesn’t satisfy my equations” a flimsy line. There is, however one interesting idea – that AI be used to police itself. A number of people are working on this and it is a good example of seeing technology realistically, as being a force for both good and bad, and that the good will triumph if we use it for human good.
This book relentlessly lays the blame at the door of AI for all kinds of injustices, but mostly it exaggerates or fails to identify the real, root causes. The book is readable, as it is lightly autobiographical, and does pose the right questions about the dangers inherent in these technologies. Unfortunately it provides exaggerated analyses and rarely the right answers. Let us remember that Weapons of Mass Destruction turned out to be lies, used to promote a disastrous war. They were sexed up through dodgy dossiers. So it is with this populist paperback.

This is an important issue being clouded by often uninformed and exaggerated. Positions. AI is unique, in my view, in having a large number of well-funded entities, set up to research and advise on the ethical issues around AI. They are doing a good job in surfacing issues, suggesting solutions and will influence regulation and policy. Hyperbolic statements based on a few flawed meme-like cases do not solve the problems that will inevitably arise. Technology is almost always a balance up upsides and downsides, let’s not throw the opportunities in education away on the basis of bias, whether in commentators or AI.

Sunday, December 17, 2017

10 uses for Chatbots in learning (with examples)

As chatbots become common in other contexts, such as retail, health and finance, so they will become common in learning. Education is always somewhat behind other sectors in considering and adopting technology but adopt it will. There are several points across the learner journey where bots are already being used and already a range of fascinating examples.
1.    Onboarding bot
Onboarding is notoriously fickle. New starters in at different times, have different needs and the old model of a huge dump of knowledge, documents and compliance courses is still all too common. Bots are being used to introduce new students or staff to the people, environment and purpose of the organisation. New starters have predictable questions, so answers can be provided straight to mobile, directed to people, processes or procedures, where necessary. It is not that the chatbot will provide the entire solution but it will take the pressure off and respond to real queries as they arise. Available 24/7 it can give access to answer as well as people. What better way to present your organization as innovative and responsive to the needs of students and staff?
2.    FAQ bot
In a sense Google is a chatbot. You type something in and up pops a set of ranked links. Increasingly you may even have a short list of more detailed questions you may want to ask. Straight up FAQ chatbots, with a well-defined set of answers to a predictable set of questions can take the load off customer queries, support desks or learner requests. A lot of teaching is admin and a chatbot can relieve that pressure at a very simple level within a definite domain – frequently asked questions.
3. Invisible LMS bot
At another level, the invisible LMS, fronted by a chatbot, allows people to ask for help and shifts formal courses into performance support, within the workflow. LearningPool’s ‘Otto’ is a good example. It sits on top of content, accessible from Facebook, Slack and other commonly used social tools. You get help in various forms, such as simple text, chunks of learning, people to contact and links to external resources as and when you need them. Content is no longer sits in a dead repository, waiting on you to sign in or take courses, but is a dynamic resource, available when you ask it something.
4. Learner engagement bot
Learners are often lazy. Students leave essays and assignments to the last minute, learners fail to do pre-work, and courses– it’s a human failing. They need prompting and cajoling. Learner engagement bots do this, with pushed prompts to students and responses to their queries. ‘Differ’ from Norway does precisely this. It recognizes that learners need to be engaged and helped, even pushed through the learning journey, and that is precisely what 'Differ' does.
5. Learner support bot
Campus support bots or course support bots go one stage further and provide teaching support in some detail. The idea is to take the administrative load off the shoulders of teachers and trainers. Response times to emails from faculty to students can be glacial. Learner support bots can, if trained well, respond with accurate and consistent answers quickly, 24/7.
The Georgia Tech bot Jill Watson, and its descendants, responds in seconds. Indeed they had to slow its response time down to mimic the typing speed of a human. The learners, 350 AI students, didn’t guess that it was a bot and even put it up for a teaching award.
6. Tutor bots
Tutor bots are different from chatbots in terms of the goals, which are explicitly ‘learning’ goals. They retain the qualities of a chatbot, flowing dialogue, tone of voice, exchange and human (like) but focus on the teaching of knowledge and skills. Straight up teaching is another approach, where the bot behaves like a Socratic teacher, asking sprints of questions and providing encouragement and feedback. This type of bot can be used as a supplement to existing courses to encourage engagement. Wildfire, the AI content generation service uses bots of this type to deliver actual teaching on apprenticeship content, as a supplement to courses, also built using AI, in minutes not months. Once the basic knowledge has been acquired, the bot tests the student as well as getting them to apply their knowledge.
7. Mentor bot
The point of a bot may not be to simply answer questions but to mentor learners by providing advice on how to find the information on your own, to promote problem solving. AutoMentor by Roger Schank,  is one such system, where the bot knows the context and provides, not just FAQ answers but advice. Providing answers is not always the best way to teach. At a higher-level chatbots could be used to encourage problem solving and critical skills, by being truly Socratic, acting as a midwife to the students behaviours and thoughts. Roger Schank is using these in defence-funded projects on Cyber Security.
As the dialogue gets better, drawing not only on a solid knowledge-base, good learner engagement through dialogue, focused and detailed feedback but also critical thought in terms of opening up perspectives, encouraging questioning of assumptions, veracity of sources and other aspects of perspectival thought, so critical thinking could also be possible. Bots will be able to analyse text to expose factual, structural or logical weaknesses. The absence of critical thought will be identified as well as suggestions for improving this skill by prompting further research ideas, sound sources and other avenues of thought. This ‘bot as critical companion’ is an interesting line of development.
8. Scenario-based bots
Beyond knowledge, we have the teaching and learning of more sophisticated scenarios, where knowledge can be applied. This is often absent in education, where almost all the effort is put into knowledge acquisition. It is easy to see why – it’s hard and time consuming. Bots can set up problems, prompt through a process, provide feedback and assess effort. Scenarios often involve other people this is where surrogate bots can come in.
9. Practice bots
Practice bots, literally take the role of a customer, patient, learners or any other person and allows learners to practice their customer care, support, healthcare or other soft skills on a responding person (bot). Bots that act as revision bots for exams are also possible.
A bot that mimics someone can be used for practice. For example, the boy with attitude ‘Eli’, developed by Penn State, that mimics an awkward child in the classroom. It is used by student teachers to practice their skills on dealing with such problems before they hit the classroom. Duolingo uses bots after you have gathered an adequate vocabulary, knowledge of grammar and basic competence, to allow practice in a language. This surely makes sense.
10. Wellbeing bots
If a bot is being used in any therapeutic context, its anonymity can be an advantage. From Eliza in the 60s to contemporary therapeutic bots, this has been a rich vein of bot development. There is an example of the word ‘suicidal’ appearing in a student messenger dialogue, that led to a fast intervention, as the student was in real distress. Therapeutic bots are being used in controlled studies to see of they have a beneficial effect on outcomes. Anonymity, in itself, is an advantage in such bots, as the learner may not want to expose their failings.
Bots such as ‘Elli ‘ and ‘Woebot’ are already being subjected to controlled trials to examine the impact on clinical outcomes.
Bot warning
The holy grail in AI is to find generic algorithms that can be used (especially in machine learning) to solve a range of different problems across a number of different domains. This is starting to happen with deep learning (machine learning). The idea is that the teacher bot will replace the skills of a teacher, not just be able to tutor in one subject alone, but be a cross-curricular teacher, especially at the higher levels of learning. It could be cross-departmental, cross-subject and cross-cultural, to produce teaching and learning that will be free from the tyranny of the institution, department, subject or culture in which it is bound. Let’s be clear, this will not happen any time soon.  AI is nowhere near solving the complex problems that this entails. If someone is promising a bot will replace a teacher – show them the door. Bots will augment not automate teaching.
We have to be careful about overreach here. Effective bots are not easy to build, have to be ‘trained (in AI-speak ‘unsupervised’) and are difficult to build. On the other hand trained bots, with good data sets (in AI-speak ‘supervised’), in specific domains, are eminently possible. Another warning is that they are on a collision course with traditional Learning Management Systems, as they usually need a dynamic server-side infrastructure. As for SCORM – the sooner it’s binned the better. Bots fit n more naturally into the xAPI landscape.
Chatbots have real potential in a number of learning activities, all along the learning journey, not as a general; ‘teacher’ but in specific applications within specific domains. They need to be trained, built, tested and improved, which is no easy task, but their efficacy in reducing the workload of teachers, trainers, lecturers and administrators is clear. The dramatic advances in Natural Language Processing have led to Siri, Amazon Echo and Google Home. It is a rapidly developing field of AI and promises to deliver chatbot technology that is better and cheaper by the month.
As a bot does not have the limitations of a human, in terms of forgetting, recall, cognitive bias, cognitive overload, getting ill, sleeping 8 hours a day, retiring and dying - once on the way to acquiring, albeit limited, skills, it will only get better and better. The more students that use its service the better it gets, not only on what it teaches but how it teaches. Courses will be fine-tuned to eliminate weaknesses, and finesse themselves to produce better outcomes.
We have seen how online behaviour has moved from flat page-turning (websites) to posting (Facebook, Twitter) to messaging (Txting, Messenger). We have seen how the web become more natural and human. As interfaces (using AI) have become more frictionless and invisible, conforming to our natural form of communication (dialogue), through text or speech. The web has become more human.
Learning takes effort. Personalised dialogue reframes learning as an exploratory, yet still structured process where the teacher guides and the learner has to make the effort. Taking the friction and cognitive load of the interface out of the equation, means the teacher and learner can focus on the task and effort needed to acquire knowledge and skills. This is the promise of bots. But the process of adoption will be gradual.

Finally, this at last is a form of technology that teachers can appreciate, as it truly tries to improve on what they already do. It takes good teaching as its standard and tries to support and streamline it to produce faster and better outcomes at a lower cost. It takes the admin and pain out of teaching. They are here, more are coming.

Thursday, December 14, 2017

7 solid reasons to suppose that chatbot interfaces will work in learning

In Raphael’s painting various luminaries stand or sit in poses on the steps, but look to the left of Plato and Aristotle and you’ll see a poor looking figure in a green robe talking to people – that’s Socrates. Most technology in teaching has run against the Socratic grain, such as the blackboard, turning teachers into talkers and lecturers. Direct instruction is necessary in one to many situations like classrooms but when the technology allows us to deliver one-to-one teaching, it makes sense to replace monologue with dialogue. With Generative AI we may be seeing the return of the Socratic method. I have written, in detail, about that method here.

This return is being enabled by Large Language Models LMS. A person is exposed to a billion words over a lifetime through speech and reading. These models have a thousand times more words, strictly speaking tokens - 1 to 2 trillion. It would take a person 22,000 years to read that at normal reading speed eight hours a day. No teacher can hole that much in long term memory.

AI is largely invisible, but it doe have to reveal itself through its user interface. AI is the new UI but because the AI is doing a lot of the smart, behind the scenes work, it is best fronted by a simple interface, the simpler the better. The dialogue interface in messenger apps, texting, and social media seems to have won the interface wars, transcending menus. Simple Socratic dialogue seems to have risen, through the process of natural selection as THE interface of choice, especially on mobile.

So can this combination of AI and Socratic UI accelerate learning? There are several reasons for being positive about this type of interface in learning.

1. Messaging the new interface
We know that dialogue interfaces overtook that of even social media, especially among the young. Look at the mobile home screen of any young person and you’ll see the dominance of chat apps. The Darwinian world of the internet is the perfect testing ground for user interfaces and messaging is what you are most likely to see when looking over the shoulder of a young person.

So one could argue that for younger audiences, chatbots are particularly appropriate, as they already use this as their main form of communication. They have certainly led the way in its use but one could also argue that there are plenty of reasons to suppose that most other people like this form of interface. It's simple and intuitive, like speaking and listening.

2. Frictionless
Easy to use, it allows you to focus on the message not the medium. The world has drifted towards dialogue for the simple reason that it is simple. By reducing the interface to its bare essentials, the learner can focus on the more important task of communications and learning. All interfaces aim to be as frictionless as possible and apart from speculative mind reading, from the likes of Elon Musk with Neuralink, this is as bare bones as one can get.

3. Reduces cognitive load
Dialogue  is simple, a radically, stripped down interface that anyone can use. It requires almost no learning and mimics what we all do in real life – simply speak and listen, then speak again. Compared to any other interface it is low on cognitive load. There is little other than a single field into which you type, it therefore goes goes at your pace.

4. Chunking
One of the joys of dialogue, and one of the reasons for its success, it that it is succinct. It is by its very nature chunked. If it were not, it wouldn’t work. Imagine being on a flight with someone, you ask them a question and get a1 hour lecture in return or imagine. Chatbots chat, they don’t talk at you.

5. Media equation
In a most likely apocryphal story, where Steve Jobs presented the Apple Mac screen to Steve Wosniak, Jobs had programmed it so say ’Hello…”. Wosniak though it uncessary – but who was right? We want our technology to be friendly, easy to use, almost our companion. This is as true on learning as it is in any other area of human endeavour.

Nass & Reeves, in The Media Equation, did 35 studies to show that we attribute agency to technology, especially computers. We anthropomorphise technology in such a way that we think the bot is human or at least exhibits human attributes. Our faculty of imagination finds this easy, as witnessed by our ready ability to suspend belief in the movies or when watching TV. It takes seconds and works in our favour with chatbots, as dialogue is a natural, evolved form of human behaviour and communication, so we are cognitively attuned to its cadence.

Geary calls speaking and listening primary skills, as they are not learnt formally. For most of our evolution we had an oral culture, unto writing arrived 5000 years ago. It is far toor recent to have had much effect on us cognitively.

6. Anonymity
If you have qualms about chat replacing human activity, remember also, that many learners are reluctant to ask their tutor, lecturer, manager or boss questions, for fear of embarrassment, as it may reveal their lack of knowledge. Others are simply quiet, even introverts. Anonymous learning, through a chatbot, then becomes a virtue not a vice. Wellbeing bots may also want to preserve anonymity. In this sense, chatbots may be superior to live, human teachers and managers. Time and time again we see how technology is preferred to human contact – ATMs, online retail and so on, in learning, in some circumstances we also witness this phenomenon.

7. Audio possible
The brain is a social organ, likes to receive stuff in chunks and interact when learning. We are social apes, grammatical geniuses at age 3 and learn to listen and speak long before we learn to read and write (which take years). Chatbots, such as Siri and Google Assistant already exist and, with the addition of text to speech and speech to text, turn chat into the exchange of speech. Reading and writing are replaced by listening and speaking. Again these are evolved abilities.

Of course, one must be careful here, as chatbots have limitations. They work best in narrow domains, with a clear purpose. Their ability to deliver full-milk, sustained dialogue has limitations. Nevertheless, they can deliver learning functions aright across the learning journey from on-boarding, learner engagement, learner support, mentoring, teaching, assessment, practice and well being.

We are about to enter the Age of AI, which is actually a return to the Age or Tracy and dialogue. It would seem that the interface itself is ideal. Of course dialogue in this context can do what no human can do, and that is deliver multimodal learning in the form of text, audio, images and video, even 3D worlds and objects. This transcends our limited bandwidth speech dialogue.

Dialogue changes ur relationship with knowledge. We no longer simply watch and listen, even search and retrieve, we engage in dialogue with knowledge. This is a different mindset. A student may get the core of a text from ChatGPT, to get over the blank page syndrome, then build, in co-creations with these tools, a full exposition or article. You learn by learning what good questions to ask, not search. This pedAIgogy is different from what has gone before, as you're speaking not to an individual teacher but a UNIVERSAL TEACHER. It pretty much knows everything that is textual, if not everything about the real world. Although, as text relates to the real world, it know some of that. It also learns from you, corrects itself as it goes.

Fully Connected by Julia Hobsbawn – I wish I hadn’t

Having seen Julia get torn to pieces by an audience in Berlin, I decided to give the book a go. But first Berlin. After an excruciating anecdote about her being in the company of Royalty in St James Palace and meeting Zak Goldsmith (it made no sense, other than name dropping), she laid out the ideas in her book describing networks as including Facebook, Ebola and Zika –all basically the same thing, a ridiculous conflation of ideas.All this social media is turning us into sheep” she bleated. Then asked “How many of you feel unhappy in your jobs?” Zero hands went up. Oh dear, try again. How many of you feel overloaded?” Three hands in a packed room. Ooops that punctured the proposition.... She then made a quick retreat into some ridiculous generalisations about her being the first to really look at networks, that Trump should be thrown off Twitter (strong anti-freedom of expression line here.... bit worrying). Basically playing the therapeutic contrarian. The audience were having none of it, many of them experts in this field.
Then came the blowback. Stephen Downes, who knows more than most on the subject of networks, was blunt “Everything you’ve said is just wrong” Wow. He then explained that there’s a large literature on networks and that the subject has been studied in depth and that she was low on knowledge and research. He was right. Andrew Keen on Stephen Downes accusation that Hobsbawn was flakey on assumptions and research "Good - glad to see someone with a hard hitting point..." Claire Fox then joined the fray.... pointing out that this contrarian stuff smacks of hysteria – it’s all a bit preachy and mumsy.
So, fast forward, I’m back from Berlin and bought the book – Fully Connected. To be fair I wanted to read the work for myself. Turns out the audience was right. 
Fully Connected
The Preface opens up with a tale about Ebola, setting the whole ‘networks are diseased and I have the cure’ tone of the book. “Culture, diseases, ideas: they’re all about networks” says Hobsbawn. Wow – she’s serious and really does want to conflate these things just to set up the self-help advice. What follows is a series of well-worn stuff about Moore’s Law, Stanley Milgram, Six Degrees of Separation, Taleb’s Black Swan, Tom Peters, Peter Drucker… punctuated by anecdotes about her and her family. It’s a curious mixture of dull, middle-class anecdotes and old school stuff without any real analysis or insights.
Ah, but here comes her insight – her new term ‘social health’. All is revealed. Her vision is pathalogocal, the usual deficit view of the modern world. All of you out there are wrapped up in evil spiders’ webs, diseased, and I have the cure. Her two big ideas are The Way to Wellbeing and The Blended Self. All of this is wrapped up in the pseudo-medical nonsense; Information obesity, Time starvation, Techno-spread, Organisational bloat. It’s like a bad diet book where you’re fed a diet of bad metaphors. Her ‘Hexagon’ of social health is the diagnosis and cure, as she puts herself forward as the next Abraham Maslow – replacing the pyramid with a hexagon – we’re networked geddit?
Part two is even worse. The usual bromides around Disconnecting, Techno-shabbats, Designing your honeycomb, The knowledge dashboard. Only then do you realise that this is a really bad, self-help book based on a few personal anecdotes and no research whatsoever.

The postscript says it all, a rambling piece about the Forth Road Bridge. I grew up in the town beneath that bridge and saw it built – but even I couldn’t see what she was on about. There are some serious writers in this area, like Andrew Keen, Nicholas Carr and others, Julia is not one of them.