Why it’s different this time

February 19, 2024February 20, 2024 Peter Cripps comment, education ai, artificial intelligence, digital skills, education, learning, mustafa suleyman, technology

Image Created Using Adobe Photoshop and Firefly

John Templeton, the American-born British stock investor, once said: “The four most expensive words in the English language are, ‘This time it’s different.’”

Templeton was referring to people and institutions who had invested in the next ‘big thing’ believing that this time it was different, the bubble could not possibly burst and their investments were sure to be safe. But then, for whatever reason, the bubble did burst and fortunes were lost.

Take as an example the tech boom of the late 1980s and 1990s. Previously unimagined technologies that no one could ever see any sign of failing meant investors poured their money into this boom. Then it all collapsed and many fortunes were lost as the Nasdaq dropped 75 percent.

It seems to be an immutable law of economics that busts will follow booms as sure as night follows day. The trick then is to predict the boom and exit your investment at the right time – not too soon and not too late, to paraphrase Goldilocks.

Most recently the phrase “this time it’s different” is being applied to the wave of AI technology which has been hitting our shores, especially since the widespread release of large language model technologies which current AI tools like OpenAI’s ChatGPT, Google’s PaLM, and Meta’s LLaMA use as their underpinning.

Which brings me to the book The Coming Wave by Mustafa Suleyman.

Suleyman was the co-founder of DeepMind (now owned by Google) and is currently CEO of Inflection an AI ‘studio’ that, according to its company blurb is “creating a personal AI for everyone”.

The Coming Wave provides us with an overview not just of the capabilities of current AI systems but also contains a warning which Suleyman refers to as the containment problem. If our future is to depend on AI technology (which it increasingly looks like it will given that, according to Suleyman, LLMs are the “fastest, diffusing consumer models we have ever seen“) how do you make it a force for good rather than evil whereby a bunch of ‘bad actors’ could imperil our very existence? In other words, how do you monitor, control and limit (or even prevent) this technology?

Suleyman’s central premise in this book is that the coming technological wave of AI is different from any that have gone before for five reasons which makes containment very difficult (if not impossible). In summary, these are:

Reason #1: Asymmetry – the potential imbalances or disparities caused by artificial intelligence systems being able to transfer extreme power from state to individual actors.
Reason #2: Exponentiality – the phenomenon where the capabilities of AI systems, such as processing power, data storage, or problem-solving ability, increase at an accelerating pace over time. This rapid growth is often driven by breakthroughs in algorithms, hardware, and the availability of large datasets.
Reason #3: Generality – the ability of an artificial intelligence system to apply their knowledge, skills, or capabilities across a wide range of tasks or domains.
Reason #4: Autonomy – the ability of an artificial intelligence system or agent to operate and make decisions independently, without direct human intervention.
Reason #5: Technological Hegemony – the malignant concentrations of power that inhibit innovation in the public interest, distort our information systems, and threaten our national security.

Suleyman’s book goes into each of these attributes in detail and I do not intend to repeat any of that here (buy the book or watch his explainer video). Suffice it to say however that collectively these attributes mean that this technology is about to deliver us nothing less than a radical proliferation of power which, if unchecked, could lead to one of two possible (and equally undesirable) outcomes:.

A surveillance state (which China is currently building and exporting).
An eventual catastrophe born of runaway development.

Other technologies have had one or maybe two of these capabilities but I don’t believe any have had all five, certainly at the level AI has. For example electricity was a general purpose technology with multiple applications but even now individuals cannot build their own generators (easily) and there is certainly not any autonomy in power generation. The internet comes closest to having all five attributes but it is not currently autonomous (though AI itself threatens to change that).

To be fair, Suleyman does not just present us with what, by any measure, is a truly wicked problem he also offers a ten point plan for for how we might begin to address the containment problem and at least dilute the effects the coming wave might have. These stretch from including built in safety measures to prevent AI from acting autonomously in an uncontrolled fashion through regulation by governments right up to cultivating a culture around this technology that treats it with caution from the outset rather than adopting the move fast and break things philosophy of Mark Zuckerberg. Again, get the book to find out more about what these measures might involve.

My more immediate concerns are not based solely on the five features described in The Coming Wave but on a sixth feature I have observed which I believe is equally important and increasingly overlooked by our rush to embrace AI. This is:

Reason #6: Techno-paralysis – the state of being overwhelmed or paralysed by the rapid pace of technological change caused by technology systems.

As is the case of the impact of the five features of Suleyman’s coming wave I see two, equally undesirable outcomes of techno-paralysis:

People become so overwhelmed and fearful because of their lack of understanding of these technological changes they choose to withdraw from their use entirely. Maybe not just “dropping out” in an attempt to return to what they see as a better world, one where they had more control, but by violently protesting and attacking the people and the organisations they see as being responsible for this “progress”. I’m talking the Tolpuddle Martyrs here but on a scale that can be achieved using the organisational capabilities of our hyper-connected world.
Rather than fighting against techno-paralysis we become irretrevably sucked into the systems that are creating and propagating these new technologies and, to coin a phrase, “drink the Kool-Aid”. The former Greek finance minister and maverick economist Yanis Varoufakis, refers to these systems, and the companies behind them, as the technofeudalists. We have become subservient to these tech overlords (i.e. Amazon, Alphabet, Apple, Meta and Microsoft) by handing over our data to their cloud spaces. By spending all of our time scrolling and browsing digital media we are acting as ‘cloud-serfs’ — working as unpaid producers of data to disproportionately benefit these digital overlords.

There is a reason why the big-five tech overlords are spending hundreds of billions of dollars between them on AI research, LLM training and acquisitions. For each of them this is the next beachhead that must be conquered and occupied, the spoils of which will be huge for those who get their first. Not just in terms of potential revenue but also in terms of new cloud-serfs captured. We run the risk of AI being the new tool of choice in weaponising the cloud to capture larger portions of our time in servitude to these companies who produce evermore ingenious ways of controlling our thoughts, actions and minds.

So how might we deal with this potentially undesirable outcome of the coming wave of AI? Surely it has to be through education? Not just of our children but of everyone who has a vested interest in a future where we control our AI and not the other way round.

Last November the UK governments Department for Education (DfE) released the results from a Call for Evidence on the use of GenAI in education. The report highlighted the following benefits:

Freeing up teacher time (e.g. on administrative tasks) to focus on better student interaction.
Improving teaching and education materials to aid creativity by suggesting new ideas and approaches to teaching.
Helping with assessment and marking.
Adaptive teaching by analysing students’ performance and pace, and to tailor educational materials accordingly.
Better accessibility and inclusion e.g. for SEND students, teaching materials could be more easily and quickly differentiated for their specific.

whilst also highlighting some potential risks including:

An over reliance on AI tools (by students and staff) which would compromise their knowledge and skill development by encouraging them to passively consume information.
Tendency of GenAI tools to produce inaccurate, biased and harmful outputs.
Potential for plagiarism and damage to academic integrity.
Danger that AI will be used for the replacement or undermining of teachers.
Exacerbation of digital divides and problems of teaching AI literacy in such a fast changing field.

I believe that to address these concerns effectively, legislators should consider implementing the following seven point plan:

Regulatory Framework: Establish a regulatory framework that outlines the ethical and responsible use of AI in education. This framework should address issues such as data privacy, algorithm transparency, and accountability for AI systems deployed in educational settings.
Teacher Training and Support: Provide professional development opportunities and resources for educators to effectively integrate AI tools into their teaching practices. Emphasize the importance of maintaining a balance between AI-assisted instruction and traditional teaching methods to ensure active student engagement and critical thinking.
Quality Assurance: Implement mechanisms for evaluating the accuracy, bias, and reliability of AI-generated content and assessments. Encourage the use of diverse datasets and algorithms to mitigate the risk of producing biased or harmful outputs.
Promotion of AI Literacy: Integrate AI literacy education into the curriculum to equip students with the knowledge and skills needed to understand, evaluate, and interact with AI technologies responsibly. Foster a culture of critical thinking and digital citizenship to empower students to navigate the complexities of the digital world.
Collaboration with Industry and Research: Foster collaboration between policymakers, educators, researchers, and industry stakeholders to promote innovation and address emerging challenges in AI education. Support initiatives that facilitate knowledge sharing, research partnerships, and technology development to advance the field of AI in education.
Inclusive Access: Ensure equitable access to AI technologies and resources for all students, regardless of their gender, socioeconomic background or learning abilities. Invest in infrastructure and initiatives to bridge the digital divide and provide support for students with special educational needs and disabilities (SEND) to benefit from AI-enabled educational tools.
Continuous Monitoring and Evaluation: Regularly monitor and evaluate the implementation of AI in education to identify potential risks, challenges, and opportunities for improvement. Collect feedback from stakeholders, including students, teachers, parents, and educational institutions, to inform evidence-based policymaking and decision-making processes.

The coming AI wave cannot be another technology that we let wash over and envelop us. Indeed Suleyman himself towards the end of his book makes the following observations…

“Technologist cannot be distant, disconnected architects of the future, listening only to themselves.“

Technologists must also be credible critics who…
“…must be practitioners. Building the right technology, having the practical means to change its course, not just observing and commenting, but actively showing the way, making the change, effecting the necessary actions at source, means critics need to be involved.“

If we are to avoid widespread techno-paralysis caused by this coming wave than we need a 21st century education system that is capable of creating digital citizens that can live and work in this brave new world.

Enchanting Minds and Machines – Ada Lovelace, Mary Shelley and the Birth of Computing and Artificial Intelligence

October 10, 2023October 10, 2023 Peter Cripps comment, education ada lovelace, alan turing, artificial intelligence, charles babbage, mary shelley, stem, turing test, women in tech

Today (10th October 2023) is Ada Lovelace Day. In this blog post I discuss why Ada Lovelace (and indeed Mary Shelley who was indirectly connected to Ada) is as relevant today as she was then.

In the summer of 1816 [1], five young people holidaying at the Villa Diodati near Lake Geneva in Switzerland found their vacation rudely interrupted by a torrential downfall which trapped them indoors. Faced with the monotony of confinement, one member of the group proposed an ingenious idea to break the boredom: each of them should write a supernatural tale to captivate the others.

Among these five individuals were some notable figures of their time. Lord Byron, the celebrated English poet and his friend and fellow poet, Percy Shelley. Alongside them was Shelley’s wife, Mary, her stepsister Claire Clairmont, who happened to be Byron’s mistress, and Byron’s physician, Dr. Polidori.

Lord Byron, burdened by the legal disputes surrounding his divorce and the financial arrangements for his newborn daughter, Ada, found it impossible to fully engage in the challenge (despite having suggested it). However, both Dr. Polidori and Mary Shelley embraced the task with fervor, creating stories that not only survived the holiday but continue to thrive today. Polidori’s tale would later appear as Vampyre – A Tale, serving as the precursor to many of the modern vampire movies and TV programmes we know today. Mary Shelley’s story, which had come to her in a haunting nightmare that very night, gave birth to the core concept of Frankenstein, published in 1818 as Frankenstein: or, The Modern Prometheus. As Jeanette Winterson asserts in her book 12 Bytes [2], Frankenstein is not just a story about “the world’s most famous monster; it’s a message in a bottle.” We’ll see why this message resounds even more today, later.

First though, we must shift our focus to another side of Lord Byron’s tumultuous life and his divorce settlement with his wife, Anabella Wentworth. In this settlement, Byron expressed his desire to shield his daughter from the allure of poetry—an inclination that suited Anabella perfectly, as one poet in the family was more than sufficient for her. Instead, young Ada received a mathematics tutor, whose duty extended beyond teaching mathematics and included eradicating any poetic inclinations Ada might have inherited. Could this be an early instance of the enforced segregation between the arts and STEM disciplines, I wonder?

Ada excelled in mathematics, and her exceptional abilities, combined with her family connections, earned her an invitation, at the age of 17, to a London soirée hosted by Charles Babbage, the Lucasian Professor of Mathematics at Cambridge. Within Babbage’s drawing room, Ada encountered a model of his “Difference Engine,” a contraption that so enraptured her, she spent the evening engrossed in conversation with Babbage about its intricacies. Babbage, in turn, was elated to have found someone who shared his enthusiasm for his machine and generously shared his plans with Ada. He later extended an invitation for her to collaborate with him on the successor to the machine, known as the “Analytical Engine”.

A Model of Charles Babbage’s *Analytical Engine*

This visionary contraption boasted the radical notion of programmability, utilising punched cards like those employed in weaving machines of that era. In 1842, Ada Lovelace (as she had become by then) was tasked with translating a French transcript of one of Babbage’s lectures into English. However, Ada went above and beyond mere translation, infusing the document with her own groundbreaking ideas about Babbage’s computing machine. These contributions proved to be more extensive and profound than the original transcript itself, solidifying Ada Lovelace’s place in history as a pioneer in the realm of computer science and mathematics.

In one of these notes, she wrote an ‘algorithm’ for the Analytical Engine to compute Bernoulli numbers, the first published algorithm (AKA computer program) ever! Although Babbage’s engine was too far ahead of its time and could not be built using current day technology, Ada is still credited as being the world’s first computer programmer. But there is another twist to this story that brings us closer to the present day.

Fast forward to the University of Manchester, 1950. Alan Turing, the now feted but ultimately doomed mathematician who led the team that cracked intercepted, coded messages sent by the German navy in WWII, has just published a paper called Computing Machinery and Intelligence [3]. This was one of the first papers ever written on artificial intelligence (AI) and it opens with the bold premise: “I propose to consider the question, ‘Can machines think?”.

Turing did indeed believe computers would one day (he thought in about 50 years’ time in the year 2000) be able to think and devised his famous “Turing Test” as a way of verifying his proposition. In his paper Turing also felt the need to “refute” arguments he thought might be made against his bold claim, including one made by no other than Ada Lovelace over one hundred years earlier. In the same notes where she wrote the world’s first computer algorithm, Lovelace also said:

“It is desirable to guard against the possibility of exaggerated ideas that might arise as to the powers of the Analytical Engine. The Analytical Engine has no pretensions whatever to originate anything. It can do whatever we know how to order it to perform. It can follow analysis, but it has no power of anticipating any analytical relations or truths”.

Although Lovelace might have been optimistic about the power of the Analytical Engine, should it ever be built, the possibility of it thinking creatively wasn’t one of the things she thought it would excel at.

Turing disputed Lovelace’s view because she could have had no idea of the enormous speed and storage capacity of modern (remember this was 1950) computers, making them a match for that of the human brain, and thus, like the brain, capable of processing their stored information to arrive at sometimes “surprising” conclusions. To quote Turing directly from his paper:

“It is a line of argument we must consider closed, but it is perhaps worth remarking that the appreciation of something as surprising requires as much of a ‘ creative mental act ‘ whether the surprising event originates from a man, a book, a machine or anything else.”

Which brings us bang up to date with the current arguments that are raging about whether systems like ChatGPT, DALL-E or Midjourney are creative or even sentient in some way. Has Turing’s prophesy finally been fulfilled or was Ada Lovelace right all along, computers can never be truly creative because creativity requires not just a reconfiguration of what someone else has made, it requires original thought based on actual human experience?

One undeniable truth prevails in this narrative: Ada was good at working with what she didn’t have. Not only was Babbage unable to build his machine, meaning Lovelace never had one to play with, she also didn’t have male privilege or a formal education – something that was a scarce commodity for women – a stark reminder of the limitations imposed on her gender during that time.

Have things moved on today for women and young girls? A glimpse into the typical composition of a computer science classroom, be it at the secondary or tertiary level, might beg the question: Have we truly evolved beyond the constraints of the past? And if not, why does this gender imbalance persist?

Over the past five or more years there have been many studies and reports published into the problem of too few women entering STEM careers and we seem to be gradually focusing in on not just what the core issues are, but also how to address them. What seems to be lacking is the will, or the funding (or both) to make it happen.

So, what to do, first some facts:

Girls lose interest in STEM as they get older. A report from Microsoft back in 2018 found that confidence in coding wanes as girls get older, highlighting the need to connect STEM subjects to real-world people and problems by tapping into girls’ desire to be creative [4].
Girls and young women do not associate STEM jobs with being creative. Most girls and young women describe themselves as being creative and want to pursue a career that helps the world. They do not associate STEM jobs as doing either of these things [4].
Female students rarely consider a career in technology as their first choice. Only 27% of female students say they would consider a career in technology, compared to 61% of males, and only 3% say it is their first choice [5].
Most students (male and female) can’t name a famous female working in technology. A lack of female role models is also reinforcing the perception that a technology career isn’t for them. Only 22% of students can name a famous female working in technology. Whereas two thirds can name a famous man [5].
Female pupils feel STEM subjects, though highly paid, are not ‘for them’. Female Key Stage 4 pupils perceived that studying STEM subjects was potentially a more lucrative choice in terms of employment. However, when compared to male pupils, they enjoyed other subjects (e.g., arts and English) more [6].

The solutions to these issues are now well understood:

Increasing the number of STEM mentors and role models – including parents – to help build young girls’ confidence that they can succeed in STEM. Girls who are encouraged by their parents are twice as likely to stay in STEM, and in some areas like computer science, dads can have a greater influence on their daughters than mums yet are less likely than mothers to talk to their daughters about STEM.
Creating inclusive classrooms and workplaces that value female opinions. It’s important to celebrate the stories of women who are in STEM right now, today.
Providing teachers with more engaging and relatable STEM curriculum, such as 3D and hands-on projects, the kinds of activities that have proven to help keep girls’ interest in STEM over the long haul.
Multiple interventions, starting early and carrying on throughout school, are important ways of ensuring girls stay connected to STEM subjects. Interventions are ideally done by external people working in STEM who can repeatedly reinforce key messages about the benefits of working in this area. These people should also be able to explain the importance of creativity and how working in STEM can change the world for the better [7].
Schoolchildren (all genders) should be taught to understand how thinking works, from neuroscience to cultural conditioning; how to observe and interrogate their thought processes; and how and why they might become vulnerable to disinformation and exploitation. Self-awareness could turn out to be the most important topic of all [8].

Before we finish, let’s return to that “message in a bottle” that Mary Shelley sent out to the world over two hundred years ago. As Jeanette Winterson points out:

“Mary Shelley maybe closer to the world that is to become than either Ada Lovelace or Alan Turing. A new kind of life form may not need to be human-like at all and that’s something that is achingly, heartbreakingly, clear in ‘Frankenstein’. The monster was originally designed to be like us. He isn’t and can’t be. Is that the message we need to hear?” [2].

If we are to heed Shelley’s message from the past, the rapidly evolving nature of AI means we need people from as diverse a set of backgrounds as possible. These should include people who can bring constructive criticism to the way technology is developed and who have a deeper understanding of what people really need rather than what they think they want from their tech. Women must become essential players in this. Not just in developing, but also guiding and critiquing the adoption and use of this technology. As Mustafa Suleyman (co-founder of DeepMind) says in his book The Coming Wave [10]:

“Credible critics must be practitioners. Building the right technology, having the practical means to change its course, not just observing and commenting, but actively showing the way, making the change, effecting the necessary actions at source, means critics need to be involved.”

As we move away from the mathematical nature of computing and programming to one driven by so called descriptive programming [9] it is going to be important we include those who are not technical but are creative as well as empathetic to people’s needs and maybe even understand the limits we should place on technology. The four C’s (creativity, critical thinking, collaboration and communications) are skills we all need to be adopting and are ones which women in particular seem to excel at.

On this, Ada Lovelace Day 2023, we should not just celebrate Ada’s achievements all those years ago but also recognize how Ada ignored and fought back against the prejudices and severe restrictions on education that women like her faced. Ada pushed ahead regardless and became a true pioneer and founder of a whole industry that did not actually really get going until over 100 years after her pioneering work. Ada, the world’s first computer programmer, should be the role model par excellence that all girls and young women look to for inspiration, not just today but for years to come.

References

Mary Shelley, Frankenstein and the Villa Diodati, https://www.bl.uk/romantics-and-victorians/articles/mary-shelley-frankenstein-and-the-villa-diodati
12 Bytes – How artificial intelligence will change the way we live and love, Jeanette Winterson, Vintage, 2022.
Computing Machinery and Intelligence, A. M. Turing, Mind, Vol. 59, No. 236. (October 1950), https://www.cs.mcgill.ca/~dprecup/courses/AI/Materials/turing1950.pdf
Why do girls lose interest in STEM? New research has some answers — and what we can do about it, Microsoft, 13th March 2018, https://news.microsoft.com/features/why-do-girls-lose-interest-in-stem-new-research-has-some-answers-and-what-we-can-do-about-it/
Women in Tech- Time to close the gender gap, PwC, https://www.pwc.co.uk/who-we-are/her-tech-talent/time-to-close-the-gender-gap.html
Attitudes towards STEM subjects by gender at KS4, Department for Education, February 2019, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/913311/Attitudes_towards_STEM_subjects_by_gender_at_KS4.pdf
Applying Behavioural Insights to increase female students’ uptake of STEM subjects at A Level, Department for Education, November 2020, https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/938848/Applying_Behavioural_Insights_to_increase_female_students__uptake_of_STEM_subjects_at_A_Level.pdf
How we can teach children so they survive AI – and cope with whatever comes next, George Monbiot, The Guardian, 8th July 2023, https://www.theguardian.com/commentisfree/2023/jul/08/teach-children-survive-ai
Prompt Engineering, Microsoft, 23rd May 2023, https://learn.microsoft.com/en-us/semantic-kernel/prompt-engineering/
The Coming Wave, Mustafa Suleyman, The Bodley Head, 2023.

Machines like us? – Part I

May 17, 2023May 30, 2023 Peter Cripps comment ai, alan turing, artificial intelligence, chatgpt, ethics, gpt, ibm, max tegmark

From *The Secret of the Machines*, Artist unknown

Our ambitions run high and low – for a creation myth made real, for a monstrous act of self love. As soon as it was feasible, we had no choice, but to follow our desires and hang the consequences.
Ian McEwan, Machines Like Me

I know what you’re thinking – not yet another post on ChatGPT! Haven’t enough words been written (or machine-generated) on this topic in the last few months to make the addition of any more completely unnecessary? What else is there to possibly say?

Well, we’ll see.

First, just in case you have been living in a cave in North Korea for the last year, what is ChatGPT? Let’s ask it…

ChatGPT is an AI language model developed by OpenAI. It is based on the GPT (Generative Pre-trained Transformer) architecture, specifically GPT-3.5. GPT-3.5 is a deep learning model that has been trained on a diverse range of internet text to generate human-like responses to text prompts.
ChatGPT response to the question: “What is ChatGPT”.

In this post, I am not interested in what use cases ChatGPT is or is not good for. I’m not even particularly interested in what jobs ChatGPT is going to replace in the coming years. Let’s face it, if the CEO of IBM, Arvind Krishna, is saying “I could easily see 30 per cent of [non-customer-facing roles] getting replaced by AI and automation over a five-year period” then many people are already going to be worried so I’m not going to add to those fears.

I see much of what Krishna predicts as inevitable. Unless the world takes note of the recent letter from the tech/AI ‘great and the good’ (which appears to have some fake signatories anyway) then the simple fact is that if you can algorithmically capture a solution to a problem, people who solve those problems for a living are probably going to get replaced by such algorithms.

What interests me is whether deep learning models like ChatGPT really are intelligent and, more significantly, are they creative. In this post I’ll consider intelligence and in a later post creativity.

It turns out that intelligence is actually quite a hard thing to define. As Max Tegmark, president of the Future of Life Institute, says: “there’s no agreement on what intelligence is even amongst intelligent intelligence researchers!” Helpfully however, in his book Life 3.0, Tegmark puts us out of our misery and defines intelligence as the: “ability to accomplish complex goals”. This, as even Tegmark acknowledges, is a very broad definition. He also claims this definition subsumes the ability to “apply knowledge and skills” as this could be a goal one sets oneself.

Having defined intelligence it then becomes relatively easy, for Tegmark at least, to define artificial intelligence (AI) as: “non biological intelligence”; general intelligence as the: “ability to accomplish virtually any goal, including learning”; artificial general intelligence (AGI) as the: “ability to accomplish any cognitive task at least as well as humans” and superintelligence as: “general intelligence far beyond human level”.

So, given the relatively low bar that these definitions are set at, where would we say ChatGPT and its ilk are currently at on the intelligence scale?

I guess this really hinges on whether you think the multitude of tasks that these deep learning models can perform constitutes “accomplishing virtually any goal” or not? For many people (especially as we’ll see, AI researchers) the answer is probably yes.

But let’s ask ChatGPT if it thinks it’s an AGI?

It is important to note that as an AI language model, I am an instantiation of artificial general intelligence (AGI) and can provide insights based on my training up until September 2021.
Partial ChatGPT response to the question: “Given Max Tegmark’s definition of intelligence where would you say you are on the various intelligence levels he proposes?”.

Personally, and up until a few weeks ago, I would have said ChatGPT was getting a bit above itself to say it was an “instantiation” of an AGI but then I read an interview with Jaron Lanier titled How humanity can defeat AI.

Lanier works for Microsoft and is the author of a number of what you might call anti-social media books including You Are Not A Gadget and Ten Arguments For Deleting Your Social Media Accounts Right Now.

Lanier’s argument in this interview is that we have got AI wrong and we should not be treating it as a new form of intelligence at all. Indeed he has previously stated there is no AI. Instead Lanier reckons we have built a new and “innovative form of social collaboration”. Like the other social collaboration platforms that Lanier has argued we should all leave because they have gone horribly wrong this new form too could become perilous in nature if we don’t design it well. In Lanier’s view therefore the sooner we understand there is no such thing as AI, the sooner we’ll start managing our new technology intelligently and learn how to use it as a collaboration tool.

Whilst all of the above is well intentioned the real insightful moment for me came when Lanier was discussing Alan Turing’s famous test for intelligence. Let me quote directly what Lanier says.

You’ve probably heard of the Turing test, which was one of the original thought-experiments about artificial intelligence. There’s this idea that if a human judge can’t distinguish whether something came from a person or computer, then we should treat the computer as having equal rights. And the problem with that is that it’s also possible that the judge became stupid. There’s no guarantee that it wasn’t the judge who changed rather than the computer. The problem with treating the output of GPT as if it’s an alien intelligence, which many people enjoy doing, is that you can’t tell whether the humans are letting go of their own standards and becoming stupid to make the machine seem smart.
Jaron Lanier, How humanity can defeat AI, UnHerd, May 8th 2023

There is no doubt that we are in great danger of believing whatever bullshit GPT’s generate. The past decade or so of social media growth has illustrated just how difficult we humans find it to handle misinformation and these new and wondrous machines are only going to make that task even harder. This, coupled with the problem that our education system seems to reward the regurgitation of facts rather than developing critical thinking skills is, as journalist Kenan Malik says, increasingly going to become more of an issue as we try to figure out what is fake and what is true.

Interestingly, around the time Lanier was saying “there is no AI”, the so called “godfather of AI”, Geoffrey Hinton was announcing he was leaving Google because he was worried that AI could become “more intelligent than humans and could be exploited by ‘bad actors'”. Clearly, as someone who created the early neural networks that were the predecessors to the large language models GPTs are built on Hinton could not be described as being “stupid”, so what is going on here? Like others before him who think AI might be exhibiting signs of becoming sentient, maybe Hinton is being deceived by the very monster he has helped create.

So what to do?

Helpfully Max Tegmark, somewhat tongue-in-cheek, has suggested the following rules for developing AI (my comments are in italics):

Don’t teach it to code: this facilitates recursive self-improvement – ChatGPT can already code.
Don’t connect it to the internet: let it learn only the minimum needed to help us, not how to manipulate us or gain power – ChatGPT certainly connected to the internet to learn what it already knows.
Don’t give it a public API: prevent nefarious actors from using it within their code – OpenAI is releasing a public API.
Don’t start an arms race: this incentivizes everyone to prioritize development speed over safety – I think it’s safe to say there is already an AI arms race between the US and China.

Oh dear, it’s not going well is it?

So what should we really do?

I think Lanier is right. Like many technologies that have gone before, AI is seducing us into believing it is something it is not – even, it seems, to its creators. Intelligent it may well be, at least by Max Tegmark’s very broad definition of what intelligence is, but let’s not get beyond ourselves. Whilst I agree (and definitely fear) AI could be exploited by bad actors it is still, at a fundamental level, little more than a gargantuan mash up machine that is regurgitating the work of the people who have written the text and created the images it spits out. These mash ups may be fooling many of us some of the time (myself included) but we must be not be fooled into losing our critical thought processes here.

As Ian McEwan points out, we must be careful we don’t “follow our desires and hang the consequences”.

Should we worry about those dancing robots?

January 16, 2021 Peter Cripps comment algorithms, artificial intelligence, ethics, robotics

The robots in question are the ones built by Boston Dynamics who shared this video over the holiday period.

For those who have not been watching the development of this companies robots, we get to see the current ‘stars’ of the BD stable, namely: ‘Atlas’ (the humanoid robot), ‘Spot’ (the ‘dog’, who else?) and ‘Handle’ (the one on wheels) all coming together for a nice little Christmassy dance.

(As an aside, if you didn’t quite get what you wanted from Santa this year, you’ll be happy to know you can have your very own ‘Spot’ for a cool $74,500.00 from the Boston Dynamics online shop).

Boston Dynamics is an American engineering and robotics design company founded in 1992 as a spin-off from the Massachusetts Institute of Technology. Boston Dynamics is currently owned by the Hyundai Motor Group (since December, 2020) having previously been owned by Google X and SoftBank Group, the Japanese multinational conglomerate holding company.

Before I get to the point of this post, and attempt to answer the question posed by it, it’s worth knowing that five years ago the US Marine Corps, working with Boston Dynamics who were under contract with DARPA, decided to abandon a project to build a “robotic mule” that would carry heavy equipment for the Marines because the Legged Squad Support System (LS3) was too noisy. I mention this for two reasons: 1) that was five years ago, a long time in robotics/AI/software development terms and 2) that was a development we were actually told about, what about all those other military projects that are classified that BD may very well be participating in? More of this later.

So back to the central question: should we worry about those dancing robots? My answer is a very emphatic ‘yes’, for three reasons.

Reason Number One: It’s a “visual lie”

The first reason is nicely summed up by James J. Ward, a privacy lawyer, in this article. Ward’s point, which I agree with, is that this is an attempt to convince people that BD’s products are harmless and pose no threat because robots are fun and entertaining. Anyone who’s been watching too much Black Mirror should just chill a little and stop worrying. As Ward says:

“The real issue is that what you’re seeing is a visual lie. The robots are not dancing, even though it looks like they are. And that’s a big problem”.

Ward goes on to explain that when we watch this video and we see these robots appearing to be experiencing the music, the rhythmic motion, the human-like gestures we naturally start to feel the joyfulness and exuberance of the dance with them. The robots become anthropomorphised and we start to feel we should love them because they can dance, just like us. This however, is dangerous. These robots are not experiencing the music or the interaction with their ‘partners’ in any meaningful way, they have simply been programmed to move in time to a rhythm. As Ward says:

“It looks like human dancing, except it’s an utterly meaningless act, stripped of any social, cultural, historical, or religious context, and carried out as a humblebrag show of technological might.”

The more content like this that we see, the more familiar and normal it seems and the more blurred the line becomes between what it is to be human and what our relationship should be with technology. In other words, we will become as accepting of robots as we are now with our mobile phones and our cars and they will suddenly be integral parts of our life just like those relatively more benign objects are.

But robots are different.

Although we’re probably still some way off from the dystopian amusement park for rich vacationers depicted in the film Westworld, where customers can live out their fantasies through the use of robots that provide anything humans want we should not ignore the threat from robots and advanced artificial intelligence (AI) too quickly. Maybe then, videos like the BD one should serve as a reminder that now is the time to start thinking about what sort of relationship we want with this new breed of machine and start developing ethical frameworks on how we create and treat things that will look increasingly like us?

Reason Number Two: The robots divert us from the real issue

If the BD video runs the risk of making us more accepting of technology because it fools us into believing those robots are just like us, it also distracts us in a more pernicious way. Read any article or story on the threats of AI and you’ll aways see it appearing alongside a picture of a robot, and usually one that Terminator like is rampaging around shooting everything and everyone in sight. The BD video however shows that robots are fun and that they’re here to do work for us and entertain us, so let’s not worry about them or, by implication, their ‘intelligence’.

As Max Tegmark points out in his book Life 3.0 however, one of the great myths of the dangers of artificial intelligence is not that robots will rise against us and wage out of control warfare Terminator style, it’s more to do with the nature of artificial intelligence itself. Namely, that an AI whose goals are misaligned with our own, needs no body, just an internet connection, to wreak its particular form of havoc on our economy or our very existence. How so?

It’s all to do with the nature of, and how we define, intelligence. It turns out intelligence is actually quite a hard thing to define (and more so to get everyone to agree on a definition). Tegmark uses a relatively broad definition:

intelligence = ability to accomplish complex goals

and it then follows that:

artificial intelligence = non-biological intelligence

Given these definitions then, the real worry is not about machines becoming malevolent but about machines becoming very competent. In other words what about if you give a machine a goal to accomplish and it decides to achieve that goal no matter what the consequences?

This was the issue so beautifully highlighted by Stanley Kubrick and Arthur C. Clarke in the film 2001: A Space Odyssey. In that film the onboard computer (HAL) on a spaceship bound for Jupiter ends up killing all of the crew but one when it fears its goal (to reach Jupiter) maybe jeopardised. HAL had no human-like manifestation (no arms or legs), it was ‘just’ a computer responsible for every aspect of controlling the spaceship and eminently able to use that power to kill several of the crew members. As far as HAL was concerned it was just achieving its goal – even if it did mean dispensing with the crew!

It seems that hardly a day goes by without there being news of not just our existing machines becoming ever more computerised but with those computers becoming ever more intelligent. For goodness sake, even our toothbrushes are now imbued with AI! The ethical question here then is how much AI is enough and just because you can build intelligence into a machine or device, does that mean you actually should?

Reason Number Three: We maybe becoming “techno-chauvinists”

One of the things I always think when I see videos like the BD one is, if that’s what these companies are showing is commercially available, how far advanced are the machines they are building, in secret, with militaries around the world?

Is there a corollary here with spy satellites? Since the end of the Cold War, satellite technology has advanced to such a degree that we are being watched — for good or for bad — almost constantly by military, and commercial organisations. Many of the companies doing the work are commercial with the boundary between military and commercial now very blurred. As Pat Norris, a former NASA engineer who worked on the Apollo 11 mission to the moon and author of Spies in the Sky says “the best of the civilian satellites are taking pictures that would only have been available to military people less than 20 years ago”. If that is so then what are the military satellites doing now?

In his book Megatech: Technology in 2050 Daniel Franklin points out that Western liberal democracies often have a cultural advantage, militarily over those who grew up under a theocracy or authoritarian regime. With a background of greater empowerment in decision making and encouragement to learn from, and not be penalised by, mistakes, Westerners tend to display greater creativity and innovation. Education systems in democracies encourage the type of creative problem-solving that is facilitated by timely intelligence as well as terabytes of data that is neither controlled nor distorted by an illiberal regime.

Imagine then how advanced some of these robots could become, in military use, if they are trained using all of the data available to them from past military conflicts, both successful and not so successful campaigns?

Which brings me to my real concern about all this. If we are training our young scientists and engineers to build ‘platforms’ (which is how Boston Dynamics refers to its robots) that can learn from all of this data, and maybe to begin making decisions which are no longer understood by their creators, then whose responsibility is it when things go wrong?

Not only that, but what happens when the technology that was designed by an engineering team for a relatively benign use, is subverted by people who have more insidious ideas for deploying those ‘platforms’? As Meredith Broussard says in her book Artificial Unintelligence: “Blind optimism about technology and an abundant lack of caution about how new technologies will be used are a hallmark of techno-chauvinism”.

As engineers and scientists who hopefully care about the future of humanity and the planet on which we live surely it is beholden on us all to morally and ethically think about the technology we are unleashing? If we don’t then what Einstein said at the advent of the atomic age rings equally true today:

“It has become appallingly obvious that our technology has exceeded our humanity.”
Albert Einstein

On Ethics and Algorithms

February 14, 2020February 18, 2020 Peter Cripps comment ai, artificial intelligence, ethics, healthcare, privacy

franck-v-g29arbbvPjo-unsplash — Photo by Franck V. on Unsplash

An article on the front page of the Observer, Revealed: how drugs giants can access your health records, caught my eye this week. In summary the article highlights that the Department of Health and Social Care (DHSC) has been selling the medical data of NHS patients to international drugs companies and have “misled” the public that the information contained in the records would be “anonymous”.

The data in question is collated from GP surgeries and hospitals and, according to “senior NHS figures”, can “routinely be linked back to individual patients’ medical records via their GP surgeries.” Apparently there is “clear evidence” that companies have identified individuals whose medical histories are of “particular interest.” The DHSC have replied by saying it only sells information after “thorough measures” have been taken to ensure patient anonymity.

As with many articles like this it is frustrating when some of the more technical aspects are not fully explained. Whilst I understand the importance of keeping their general readership on board and not frightening them too much with the intricacies of statistics or cryptography it would be nice to know a bit more about how these records are being made anonymous.

There is a hint of this in the Observer report when it states that the CPRD (the Clinical Practice Research Datalink ) says the data made available for research was “anonymous” but, following the Observer’s story, it changed the wording to say that the data from GPs and hospitals had been “anonymised”. This is a crucial difference. One of the more common methods of ‘anonymisation’ is to obscure or redact some bits of information. So, for example, a record could have patient names removed and ages and postcodes “coarsened”, that is only the first part of a postcode (e.g. SW1A rather than SW1A 2AA) are included and ages are placed in a range rather than using someones actual age (e.g. 60-70 rather than 63).

The problem with anonymising data records is that they are prone to what is referred to as data re-identification or de-anonymisation. This is the practice of matching anonymous data with publicly available information in order to discover the individual to which the data belongs. One of the more famous examples of this is the competition that Netflix organised encouraging people to improve its recommendation system by offering a $50,000 prize for a 1% improvement. The Netflix Prize was started in 2006 but abandoned in 2010 in response to a lawsuit and Federal Trade Commission privacy concerns. Although the dataset released by Netflix to allow competition entrants to test their algorithms had supposedly been anonymised (i.e. by replacing user names with a meaningless ID and not including any gender or zip code information) a PhD student from the University of Texas was able to find out the real names of people in the supplied dataset by cross-referencing the Netflix dataset with Internet Movie Database (IMDB) ratings which people post publicly using their real names.

Herein lies the problem with the anonymisation of datasets. As Michael Kearns and Aaron Roth highlight in their recent book The Ethical Algorithm, when an organisation releases anonymised data they can try and make an intelligent guess as to which bits of the dataset to anonymise but it can be difficult (probably impossible) to anticipate what other data sources either already exist or could be made available in the future which could be used to correlate records. This is the reason that the computer scientist Cynthia Dwork has said “anonymised data isn’t” – meaning either it isn’t really anonymous or so much of the dataset has had to be removed that it is no longer data (at least in any useful way).

So what to do? Is it actually possible to release anonymised datasets out into the wild with any degree of confidence that they can never be de-anonymised? Thankfully something called differential privacy, invented by the aforementioned Cynthia Dwork and colleagues, allows us to do just that. Differential privacy is a system for publicly sharing information about a dataset by describing the patterns of groups within the dataset while withholding information about individuals in that dataset.

To understand how differential privacy works consider this example*. Suppose we want to conduct a poll of all people in London to find out who have driven after taking non-prescription drugs. One way of doing this is to randomly sample a suitable number of Londoners, asking them if they have ever driven whilst under the influence of drugs. The data collected could be entered into a spreadsheet and various statistics, e.g. number of men, number of women, maybe ages etc derived. The problem is that whilst collecting this information lots of compromising personal details may be collected which, if the data were stolen, could be used against them.

In order to avoid this problem consider the following alternative. Instead of asking people the question directly, first ask them to flip a coin but not to tell us how it landed. If the coin comes up heads they tell us (honestly) if they have driven under the influence. If it comes up tails however they tell us a random answer then flip the coin again and tell us “yes” if it comes up heads or “no” if it is tails. This polling protocol is a simple randomised algorithm which is a form of differential privacy. So how does this work?

When we ask people if they have driven under the influence using this protocol half the time (i.e. when the coin lands heads up) the protocol tells them to tell the truth. If the protocol tells them to respond with a random answer (i.e. when the coin lands tails up), then half of that time they just happen to randomly tell us the right answer. So they tell us the right answer 1/2 + ((1/2) x (1/2)) or three-quarters of the time. The remaining one quarter of the time they tell us a lie. There is no way of telling true answers from lies. Surely though, this injection of randomisation completely masks the true results and the data is now highly error prone? Actually, it turns out, this is not the case.

Because we know how this randomisation is introduced we can reverse engineer the answers we get to remove the errors and get an approximation of the right answer. Here’s how. Suppose one-third of people in London have actually driven under the influence of drugs. So of the one-third who have truthfully answered “yes” to the question, three-quarters of those will answer “yes” using the protocol, that is 1/3 x 3/4 = 1/4. Of the two-thirds who have a truthful answer of “no”, one-quarter of those will report “yes”, that is 2/3 x 1/4 = 1/6. So we expect 1/4 + 1/6 = 5/12 ~ 1/3 of the population to answer “yes”.

So what is the point of doing the survey like this? Simply put it allows the true answer to be hidden behind the protocol. If the data were leaked and an individual from it was identified as being suspected of driving under the influence then they could always argue they were told to say “yes” because of the way the coins fell.

In the real world a number of companies including the US census, Apple, Google and Privitar Lens use differential privacy to limit the disclosure of private information about individuals whose information is in public databases.

It would be nice to think that the NHS data that is supposedly being used by US drug companies was protected by some form of differential privacy. If it were, and if this could be explained to the public in a reasonable and rational way, then surely we would all benefit both in the knowledge that our data is safe and is maybe even being put to good use in protecting and improving our health. After all, wasn’t this meant to be the true benefit of living in a connected society where information is shared for the betterment of all our lives?

*Based on an example from Kearns and Roth in The Ethical Algorithm.

Cummings needs data scientists, economists and physicists (oh, and weirdos)

January 8, 2020 Peter Cripps comment ai, artificial intelligence, dominic cummings, uk government

To answer my (rhetorical) question in this post I think it’s been pretty much confirmed since the election that Dominic Cummings is, in equal measures, the most influential, disruptive, powerful and dangerous man in British politics right now. He has certainly set the cat amongst the pigeons in this blog post where he has effectively by-passed the civil service recruitment process by advertising for people to join his ever growing team of SPAD’s (special advisors). Cummings is looking for data scientists, project managers, policy experts and assorted weirdos to join his team. (Interestingly today we hear that the self-proclaimed psychic Uri Geller has applied for the job believing he qualifies because of the super-talented weirdo aspect of the job spec.)

Cummings is famed for his wide reaching reading tastes and the job spec also cites a number of scientific papers potential applicants “will be considering”. The papers mentioned are broadly in the areas of complex systems and the use of maths and statistics in forecasting which give an inkling into the kind of problems Cummings sees as those that need to be ‘fixed’ in the civil service as well as the government at large (including the assertion that “Brexit requires many large changes in policy and in the structure of decision-making”).

Like many of his posts, this particular one tends to ramble and also be contradictory. In one paragraph he’s saying that you “do not need a PhD” but then in the very next one saying you “must have exceptional academic qualifications from one of the world’s best universities with a PhD or MSc in maths or physics.”

Cummings also returns to one of his favourite topics which is that of the failure of projects – mega projects in particular – and presumably those that governments tend to initiate and not complete on time or to budget (or at all). He’s an admirer of some of the huge project successes of yesteryear such as The Manhattan Project (1940s), ICBMs (1950s) and Apollo (1960s) but reckons that since then the Pentagon has “systematically de-programmed itself from more effective approaches to less effective approaches from the mid-1960s, in the name of ‘efficiency’.” Certainly the UK government is no stranger to some spectacular project failures itself both in the past and present (HS2 and Crossrail being two more contemporary examples of not so much failures but certainly massive cost overruns).

However as John Naughton points out here “these inspirational projects have some interesting things in common: no ‘politics’, no bureaucratic processes and no legal niceties. Which is exactly how Cummings likes things to be.” Let’s face it both Crossrail and HS2 would be a doddle of only you could do away with all those pesky planning proposals and environmental impact assessments you have to do and just move people out of the way quickly – sort of how they do things in China maybe?

Cummings believes that now is the time to bring together the right set of people with a sufficient amount of cognitive diversity and work in Downing Street with him and other SPADs to start to address some of the wicked problems of government. One ‘lucky’ person will be his personal assistant, a role which he says will “involve a mix of very interesting work and lots of uninteresting trivia that makes my life easier which you won’t enjoy.” He goes on to say that in this role you “will not have weekday date nights, you will sacrifice many weekends — frankly it will hard having a boy/girlfriend at all. It will be exhausting but interesting and if you cut it you will be involved in things at the age of ~21 that most people never see.” That’s quite some sales pitch for a job!

What this so called job posting is really about though is another of Cummings abiding obsessions (which he often discusses in his blog) that the government in general, and civil service in particular (which he groups together as “SW1”), is basically not fit for purpose because it is scientifically and technologically illiterate as well as being staffed largely with Oxbridge humanities graduates. The posting is also a thinly veiled attempt at pushing the now somewhat outdated ‘move fast and break things” mantra of Silicon Valley. An approach that does not always play out well in government (Universal Credit anyone). I well remember my time working at the DWP (yes, as a consultant) where one of the civil servants with whom I was working said that the only problem with disruption in government IT was that it was likely to lead to riots on the streets if benefit payments were not paid on time. Sadly, Universal Credit has shown us that it’s not so much street riots that are caused but a demonstrable increase in demand for food banks. On average, 12 months after roll-out, food banks see a 52% increase in demand, compared to 13% in areas with Universal Credit for 3 months or less.

Cummings of course would say that the problem is not so much that disruption per se causes problems but rather the ineffective, stupid and incapable civil servants who plan and deploy such projects are at fault, hence the need for hiring the right ‘assorted weirdos’ who will bring new insights that fusty old civil servants cannot see. Whilst he may well be right that SW1 is lacking in deep technical experts as well as great project managers and ‘unusual’ economists he needs to realise that government transformation cannot succeed unless it is built on a sound strategy and good underlying architecture. Ideas are just thoughts floating in space until they can be transformed into actions that result in change which takes into account that the ‘products’ that governments deal with are people not software and hardware widgets.

This problem is far better articulated by Hannah Fry when she says that although maths has, and will continue to have, the capability to transform the world those who apply equations to human behaviour fall into two groups: “those who think numbers and data ultimately hold the answer to everything, and those who have the humility to realise they don’t.”

Possibly the last words should be left to Barack Obama who cautioned Silicon Valley’s leaders thus:

“The final thing I’ll say is that government will never run the way Silicon Valley runs because, by definition, democracy is messy. This is a big, diverse country with a lot of interests and a lot of disparate points of view. And part of government’s job, by the way, is dealing with problems that nobody else wants to deal with.

So sometimes I talk to CEOs, they come in and they start telling me about leadership, and here’s how we do things. And I say, well, if all I was doing was making a widget or producing an app, and I didn’t have to worry about whether poor people could afford the widget, or I didn’t have to worry about whether the app had some unintended consequences — setting aside my Syria and Yemen portfolio — then I think those suggestions are terrific. That’s not, by the way, to say that there aren’t huge efficiencies and improvements that have to be made.

But the reason I say this is sometimes we get, I think, in the scientific community, the tech community, the entrepreneurial community, the sense of we just have to blow up the system, or create this parallel society and culture because government is inherently wrecked. No, it’s not inherently wrecked; it’s just government has to care for, for example, veterans who come home. That’s not on your balance sheet, that’s on our collective balance sheet, because we have a sacred duty to take care of those veterans. And that’s hard and it’s messy, and we’re building up legacy systems that we can’t just blow up.”

Now I think that’s a man who shows true humility, something our current leaders (and their SPADs) could do with a little more of I think.

From Turing to Watson (via Minsky)

January 27, 2016 Peter Cripps comment alan turing, artificial intelligence, cognitive computing, ibm, marvin minsky, watson

This week (Monday 25th) I gave a lecture about IBM’s Watson technology platform to a group of first year students at Warwick Business School. My plan was to write up the transcript of that lecture, with links for references and further study, as a blog post. The following day when I opened up my computer to start writing the post I saw that, by a sad coincidence, Marvin Minsky the American cognitive scientist and co-founder of the Massachusetts Institute of Technology’s AI laboratory had died only the day before my lecture. Here is that blog post, now updated with some references to Minsky and his pioneering work on machine intelligence.

First though, let’s start with Alan Turing, sometimes referred to as “the founder of computer science”, who led the team that developed a programmable machine to break the Nazi’s Enigma code, which was used to encrypt messages sent between units on the battlefield during World War 2. The work of Turing and his team was recently brought to life in the film The Imitation Game starring Benedict Cumberbatch as Turing and Keira Knightley as Joan Clarke, the only female member of the code breaking team.

Sadly, instead of being hailed a hero, Turing was persecuted for his homosexuality and committed suicide in 1954 having undergone a course of hormonal treatment to reduce his libido rather than serve a term in prison. It seems utterly barbaric and unforgivable that such an action could have been brought against someone who did so much to affect the outcome of WWII. It took nearly 60 years for his conviction to be overturned when on 24 December 2013, Queen Elizabeth II signed a pardon for Turing, with immediate effect.

In 1949 Turing became Deputy Director of the Computing Laboratory at Manchester University, working on software for one of the earliest computers. During this time he worked in the emerging field of artificial intelligence and proposed an experiment which became known as the Turing test having observed that: “a computer would deserve to be called intelligent if it could deceive a human into believing that it was human.”

The idea of the test was that a computer could be said to “think” if a human interrogator could not tell it apart, through conversation, from a human being.

Turing’s test was supposedly ‘passed’ in June 2014 when a computer called Eugene fooled several of its interrogators that it was a 13 year old boy. There has been much discussion since as to whether this was a valid run of the test and that the so called “supercomputer,” was nothing but a chatbot or a script made to mimic human conversation. In other words Eugene could in no way considered to be intelligent. Certainly not in the sense that Professor Marvin Minsky would have defined intelligence at any rate.

In the early 1970s Minsky, working with the computer scientist and educator Seymour Papert, wrote a book called The Society of Mind, which combined both of their insights from the fields of child psychology and artificial intelligence.

Minsky and Papert believed that there was no real difference between humans and machines. Humans, they maintained, are actually machines of a kind whose brains are made up of many semiautonomous but unintelligent “agents.” Their theory revolutionized thinking about how the brain works and how people learn.

Despite the more widespread accessibility to apparently intelligent machines with programs like Apple Siri Minsky maintained that there had been “very little growth in artificial intelligence” in the past decade, saying that current work had been “mostly attempting to improve systems that aren’t very good and haven’t improved much in two decades”.

Minsky also thought that large technology companies should not get involved the field of AI saying: “we have to get rid of the big companies and go back to giving support to individuals who have new ideas because attempting to commercialise existing things hasn’t worked very well,”

Whilst much of the early work researching AI certainly came out of organisations like Minsky’s AI lab at MIT it seems slightly disingenuous to believe that commercialistion of AI, as being carried out by companies like Google, Facebook and IBM, is not going to generate new ideas. The drive for commercialisation (and profit), just like war in Turing’s time, is after all one of the ways, at least in the capitalist world, that innovation is created.

Which brings me nicely to Watson.

IBM Watson is a technology platform that uses natural language processing and machine learning to reveal insights from large amounts of unstructured data. It is named after Thomas J. Watson, the first CEO of IBM, who led the company from 1914 – 1956.

IBM Watson was originally built to compete on the US television program Jeopardy. On 14th February 2011 IBM entered Watson onto a special 3 day version of the program where the computer was pitted against two of the show’s all-time champions. Watson won by a significant margin. So what is the significance of a machine winning a game show and why is this a “game changing” event in more than the literal sense of the term?

Today we’re in the midst of an information revolution. Not only is the volume of data and information we’re producing dramatically outpacing our ability to make use of it but the sources and types of data that inform the work we do and the decisions we make are broader and more diverse than ever before. Although businesses are implementing more and more data driven projects using advanced analytics tools they’re still only reaching 12% of the data they have, leaving 88% of it to go to waste. That’s because this 88% of data is “invisible” to computers. It’s the type of data that is encoded in language and unstructured information, in the form of text, that is books, emails, journals, blogs, articles, tweets, as well as images, sound and video. If we are to avoid such a “data waste” we need better ways to make use of that data and generate “new knowledge” around it. We need, in other words, to be able to discover new connections, patterns, and insights in order to draw new conclusions and make decisions with more confidence and speed than ever before.

For several decades we’ve been digitizing the world; building networks to connect the world around us. Today those networks connect not just traditional structured data sources but also unstructured data from social networks and increasingly Internet of Things (IoT) data from sensors and other intelligent devices.

These additional sources of data mean that we’ve reached an inflection point in which the sheer volume of information generated is so vast; we no longer have the ability to use it productively. The purpose of cognitive systems like IBM Watson is to process the vast amounts of information that is stored in both structured and unstructured formats to help turn it into useful knowledge.

There are three capabilities that differentiate cognitive systems from traditional programmed computing systems.

Understanding: Cognitive systems understand like humans do, whether that’s through natural language or the written word; vocal or visual.
Reasoning: They can not only understand information but also the underlying ideas and concepts. This reasoning ability can become more advanced over time. It’s the difference between the reasoning strategies we used as children to solve mathematical problems, and then the strategies we developed when we got into advanced math like geometry, algebra and calculus.
Learning: They never stop learning. As a technology, this means the system actually gets more valuable with time. They develop “expertise”. Think about what it means to be an expert- – it’s not about executing a mathematical model. We don’t consider our doctors to be experts in their fields because they answer every question correctly. We expect them to be able to reason and be transparent about their reasoning, and expose the rationale for why they came to a conclusion.

The idea of cognitive systems like IBM Watson is not to pit man against machine but rather to have both reasoning together. Humans and machines have unique characteristics and we should not be looking for one to supplant the other but for them to complement each other. Working together with systems like IBM Watson, we can achieve the kinds of outcomes that would never have been possible otherwise:

IBM is making the capabilities of Watson available as a set of cognitive building blocks delivered as APIs on its cloud-based, open platform Bluemix. This means you can build cognition into your digital applications, products, and operations, using any one or combination of a number of available APIs. Each API is capable of performing a different task, and in combination, they can be adapted to solve any number of business problems or create deeply engaging experiences.

So what Watson APIs are available? Currently there are around forty which you can find here together with documentation and demos. Four examples of the Watson APIs you will find at this link are:

Dialog

Use natural language to automatically respond to user questions

Visual Recognition

Analyses the contents of an image or video and classifies by category.

Text to Speech

Synthesize speech audio from an input of plain text.

Personality Insights

Understand someones personality from what they have written.

It’s never been easier to get started with AI by using these cognitive building blocks. I wonder what Turing would have made of this technology and how soon someone will be able to pin together current and future cognitive building blocks to really pass Turing’s famous test?

Software Architecture Zen

Peter Cripps' thoughts, guidance and best practice on the field of software architecture.

artificial intelligence