The Charybdis is a swirling water feature in the temperate house at Savill Garden. It was designed by Giles Rayner in 2006. https://funandgames.org/web/wp-content/uploads/2020/09/The-Charybdis_Savill-Gardens_9257-2-scaled.jpg– you may have recently seen the original in The Odyssey
Last week I mentioned that I thought this blog, for the rest of the year at least, would be about what we resist, what we create and what we could nurture in the face of a terrifying future. This week I thought I should start to put a bit of flesh on that particular carcass.
One source for my thinking on this is Alex Steffen. I recently attended one of his Personal Climate Strategy: The Basics webinars. He is particularly good on the importance of continually updating your world view in the face of climate discontinuity and increasing evidence of societal brittleness, and how professions can often be a barrier to doing this:
The greatest danger in any work that asks you to think systemically about the future is getting locked into the worldview that made sense to you when you first began, that you built your successful career on.
We all have limited time and energy. Building up an insightful mental model of how the world works takes a lot of both. The pay-off is in the profit and sense of purpose gained from one’s expertise. It is very common, when you’re highly rewarded for a given set of working insights, to commit more to those insights as your career unfolds, to begin even to defend those insights from challenging new perspectives (ones you fear might devalue your intellectual stock in trade). This “sunk-cost expertise” can easily become a set of shackles.
The article then goes on to explain the process of triangulation, by which professionals can “deflate pressure (from the public, regulators or activist stakeholders) to change more quickly and shore up the power of management and experts who would otherwise be forced to admit that they don’t really know what they’re doing”. If someone starts by saying that they support what you are suggesting, then responds with a bunch of short-term easy actions and claims that these are the start of a journey towards what you are suggesting, you are probably being triangulated.
The biological idea that first made me think this might be possible is called allostasis. It is not a complicated concept. It simply describes the way living organisms survive by continually adapting to the changing world around them.
As we all know, nothing alive exists in a stable environment. Every living thing must continually respond and adapt to changing circumstances. Human beings are no exception. Our brains and bodies are constantly anticipating change and adjusting to it. Most of that work happens without us ever noticing. Allostasis happens all the time, and most of us do not notice. But the fact is that this continual adaptation has a cost.
Murphy points out that, at a time of continual change like ours, that cost is high. And the adaptation required is much higher for people who cannot just throw money at their problems. He then goes on to say:
That changes how we think about political economy. Politics then becomes the process by which societies decide how those costs of adaptation are distributed, whilst economics becomes the study of how institutions either increase or reduce them. Accounting becomes the means by which we identify who ultimately bears the cost of adaptation and health becomes one of the principal ways in which we observe its effects.
I think there is a lot to this approach, which would also provide a structure to identify and critique the triangulation process set out by Steffen. Murphy only mentions education once in this piece:
Education should recognise that human beings learn differently, rather than assuming that everyone should conform to a single model.
I would go a lot further than that, because I think education is the key to helping individuals with their own adaptation strategies as well as in developing the political understanding to see how the costs of adaptation are currently distributed and how to change that within a democratic system. This seems to be an extension of what I have been writing recently about how not to be a reverse-centaur for instance.
The implications for this in health, education and our institutions are considerable in my view and growing all the time. So that is what I am going to be writing about for a while.
Isaac Newton may or may not have been nutted by an apple. His friend William Stukeley, whose memoir of Newton was the source of the story, states it as follows (spelling and punctuation from Stukeley’s manuscript at the Royal Society):
“why should that apple always descend perpendicularly to the ground,” thought he to him self: occasion’d by the fall of an apple, as he sat in a comtemplative mood: “why should it not go sideways, or upwards? but constantly to the earths centre? assuredly, the reason is, that the earth draws it. there must be a drawing power in matter. & the sum of the drawing power in the matter of the earth must be in the earths center, not in any side of the earth. therefore dos this apple fall perpendicularly, or toward the center. if matter thus draws matter; it must be in proportion of its quantity. therefore the apple draws the earth, as well as the earth draws the apple.”
However there may be stronger grounds for believing that Newton nutted English mathematics and, as a result, the actuarial profession.
I thought all this to myself, occasion’d by participation in the Actuarial Teachers and Researchers Conference (ATRC) this week. It was great. There were lots of interesting talks, and people were very engaged and respectful in how they discussed them. There was a lot of expertise in the room about both education and actuarial practice and theory. It felt like it was a room that could take on difficult topics and make progress in tackling them.
There were several calls for the need to change assessment, one of them by me, whether due to large student cohorts and the difficulties of engaging them, or the ease with which AI can duplicate examination solutions or because the credentials provided by those assessments seem increasingly irrelevant to the actual actionable skills, knowledge and experience required to operate successfully as an actuary. And there was an equally robust response from the Institute and Faculty of Actuaries (IFoA) to the effect that assessments would not be changing significantly in the latest changes due to come in by 2029.
My favourite presentation was by Angus Macdonald, about his recent short paper on Newton, Leibniz and Actuarial Science. It conjectures that the argument between Leibniz and Newton over who invented calculus resulted in the stagnation of mathematics in English universities and meant that actuaries needing to gain respectability for their advice were forced to create a professional body (in company with architects, accountants and engineers) rather than rely on universities to develop actuarial thinking as happened in continental Europe. This stagnation (Macdonald quoted G.H.Hardy, talking about the Cambridge mathematics exams, saying that they had “. . . effectively ruined serious mathematics in England for a hundred years”) was felt across the whole Anglosphere, with attempts to create institutions with equal status to the traditional universities leading to the establishment of business schools in the USA in the late 19th and early 20th centuries.
This explained a lot for me. The suspicion of our actuarial courses amongst some of our mathematics colleagues at times at Leicester. The occasionally uneasy relationship between the IFoA and the accredited universities where students can gain exemptions from their exams. The horror from some at the very idea that an actuary might qualify as a result of university courses, like they can in the Netherlands, for instance. ATRC seemed the appropriate place to explore these ideas.
Angus Macdonald considers that the Newtonian and Leibnizian branches have nearly rejoined, but I feel that our actuarial education system still suffers from the long shadow cast by these two 17th century gentlemen and their personal enmity. The ferocity with which some within the profession opposed what they saw as a threat to the popularity of the fellowship qualification posed by the new chartered actuary designation was hard to explain when the number of practising certificates issued for chief actuaries, pension scheme actuaries, etc was only around 1,100 (out of over 17,000 fellows), but if the whole profession originated from a highly developed form of status anxiety compared to academics, it becomes much easier to understand. As they say, the apple doesn’t fall far from the tree.
Now, nearly 180 years on from this “unstoppable” “rush to respectabilize” (according to Jeremy Paxman) I think it is legitimate to ask:
Why does the IFoA still insist on detailed accreditation of individual university courses on basic mathematics, statistics and business economics? Do they not trust them to teach them right? Do they think that there is something better about learning these universal subjects via a bespoke course created by a tiny professional body with 34,000 members globally?
Why are they so concerned about what assessment methods the universities use? Do they consider that they have more educational expertise on this than some of the leading universities in the world?
I have written previously about the changes in assessment I think are necessary and floated some ideas about where actuarial education might sit to accommodate these. My view was that the IFoA, faced with the need to innovate at all levels of its education system at a time of great uncertainty, might wish to get out of the foundational mathematics and business education tuition of the core principles subjects and leave this to the university system, in a truly Leibnizian way.
In actuarial science, the list of UK universities offering accredited courses has not changed for some time, and it is a small group within the university system. These courses must be seen as vulnerable. I don’t think it is a coincidence that it is so hard to persuade one of this tiny band to run the ATRC. This week’s was the first since 2021 and the list of past hosts is not very long:
The IFoA needs to ask itself what it would do if there were some high profile closures of actuarial courses and whether perhaps it might be time to protect the core mathematical education arrangements for actuaries through a broad brush accreditation of a wider range of courses, rather than the very detailed accreditation process it currently carries out on core principles subjects.
I also doubt that it has the capacity on its own to significantly restructure its assessments beyond the objective based assessments it has developed to work within its proctored environment.
The core practice and specialist subjects leading to fellowship are the areas where actuaries have, rightly, the strongest views about content and assessment structures. It surely makes sense that the expertise within the profession and the university system are pooled in a joint endeavour rather than risking both failing independently.
In contrast to Newton, who had quite a temper, Leibniz was famously parodied by Voltaire in his Dr Pangloss character in Candide who believed that “All is for the best in this best of all possible worlds.”
Let’s not be too Panglossian about the current state of actuarial education.
As I discussed in How Not To Be A Reverse-Centaur (Cory Doctorow defines a reverse-centaur as “a machine that is assisted by a human being, who is expected to work at the machine’s pace”) our actuarial education system needs to change. This has been true for some time, but the development of AI systems has both demonstrated why more clearly and accelerated the timeframes over which action is needed. As I said in December last year:
A large part of the education of the future will need to be about equipping us all to understand what we now have access to and when and how to access it. We will all have different things we are interested in, or end up involved with and needing to be educated about. It will be up to each of us to decide which things are worth the difficulty of learning, but to make those decisions we will need education that can support the development of judgement.
For education institutions, the question will be what is not worth the difficulty of learning? Credentialising based on now relatively meaningless assessment methods will not cut it. This is where the confrontation with employers and politicians is likely to come. Essential skills and their related knowledge will be better developed and assessed via more open-ended project work and online assessment of it to check understanding. These will need to become the norm, with written examinations becoming less and less prevalent. Not because of fear of cheating and plagiarism, but because an outcome which can be replicated that easily by AI is not worth assessing in the first place.
Suppose you were to ask someone who had never seen our academic system that, in order to assess whether someone else we wanted them to employ in their organisation would do a good job for them, we would:
Award marks for what has been written in answer to academic questions about what the students can remember unaided about the content of their lecture courses and reading lists with a biro on a pad of paper perched precariously on a tiny wooden table surrounded by hundreds of other similar scribblers, for a set period of time as minders wandered the floors like Victorian factory owners.
These marks would be based on marking criteria they would never see.
We would then get together in a secret huddle for a couple of months to mark all these scripts, check them and “moderate” them to check they were in line with the previous years’ exercises.
Then out would come the answer (like the Answer of 42 to the Ultimate Question of Life, the Universe and Everything in Douglas Adams’ The Hitchhikers Guide To The Galaxy).
Imagine another world where the kinds of things employers routinely do when trying to decide whether to take on someone as a new member of their organisation are routinely part of their assessments at university, eg:
Watch them operating in groups working on a task.
Ask them to describe how they would go about tackling a particular problem.
Get them to show they understand the implications of a piece of work they have carried out and can explain what assumptions it is based on and can justify them.
With assessments of performance which are not opaque academic exercises but descriptions of performance made as the tasks are being carried out evidenced by video if necessary. In other words, transparent enough to whoever might want to work with these students next to eliminate the gap between their credentials and and their actual actionable skills, knowledge and experience.
Assessing in this way will be highly demanding, for both the students and the assessors, but it means the final assessment is done with the student present and, with careful probing from the assessors, who will obviously need to have done a close reading of the project work beforehand, confidence will be high about the abilities demonstrated by the process. And it removes the need for all the time currently spent on the Victorian factory owner process – imagine what could be done if the whole of May, June and July wasn’t spent marking scripts!
A move like this will significantly change what is taught. There will be no point presenting lots of information in ways which make retention easier for students (the so-called bookwork” questions currently in most exams). This makes sense when you think about how much of the content of your courses is accessible to you now, even on a vocational course like actuarial science, as opposed to skills developed. Instead the focus will be more on conversations with students for the purpose of developing their understanding of a meaty topic and the problems to be solved within it. Explorations of these will form the projects students will be spending most of their time working on. Checking understanding and making sure students are working within frameworks they understand and can talk about will be the main teaching activity, with opportunities to practise these throughout the year. In other words, showing the ability to apply what they have understood and demonstrate the ability to reflect, make judgements and show higher levels of understanding as a result. An ability they can then apply to new problems and situations.
And notice something else about how these alternative credentials are generated? They are all social activities: from operating in groups, to presenting a piece of work to a group of assessors. Even the project work is developed through group sessions alongside individual work. If you want to develop people who can spot the gaps between models and reality you need to do it in groups, where people can test their perspectives against each other.
Compare this to the current solitary exam preparation, peppered with “revision” lectures, exam sitting and exam marking processes which comprise the predominant activities in universities between March and July each year. We are currently being funneled into an ever more solitary professional practice around LLMs. It has become quite accepted for people to do all of their thinking on a subject closeted with perhaps one, two or three LLMs but no other people and then share their analysis with an expectation that other people should read the output. Developing the skills to do these analyses will obviously be necessary, but they will never be the most important skills people learn.
Treating each other like walking databases is never going to make you new friends or influence people. However the skills you develop via an alternative more social education will.
There are examples of this approach already happening, with employers involved in the design of assessments in some cases and universities using the flexibility they have to restructure and redesign courses. But the dominant assessment system of formal exams is keeping much of this activity at the margins at the moment.
If it were to become the dominant assessment, it would also mean that larger areas of the curriculum could be examined, as the focus would not be on retention of large bodies of content but the ability to use the content to solve problems. So, perhaps, after a first year levelling up students’ ability and experience of mathematics and statistics, year 2 would have an actuarial statistics module (currently CS1/CS2) and an actuarial mathematics module (currently CM1/CM2), both dissertation based and vivaed. Then in the third year they would tackle an economics module and a business modules (currently CB1 and CB3). Perhaps in the final year of a four year MAct they would tackle a modelling and communication module (CP2/CP3) and either CP1 actuarial practice module would be offered or a professional skills module.
This would be more of a driving test style of assessment, with students working with supervisors on the comments from assessors on the original dissertations and presentations until they were up to the required standard. Students unable to make the standard after an additional year would be offered a move to a non-accredited course. The current funding model for higher education will need to change: loading students with £10,000 more debt every time they add to the nation’s skills bank with another year wrestling with these difficult skills to master was never a clever strategy, but this structure would make it even clearer how self defeating it is. If economists are accepting that the labour market no longer allocates resources effectively and are considering either a universal basic income or universal basic services, then logically university funding should follow suit.
The implications for the actuarial profession will be even more challenging. I question whether the Institute and Faculty of Actuaries will continue to consider it worth the effort and expense of reconfiguring their entire approach to exam setting, syllabus maintenance, marking, online proctoring, etc for subjects routinely taught throughout the university system for a a variety of purposes, ie the current core principles subjects of mathematics, statistics and business studies.
Whether they retain their own capacity to assess the core practice subjects is a more open question (CP1 Actuarial Practice, I imagine, will be seen as sufficiently specialist to keep in house, with perhaps a few universities, as now, accredited to run courses which can earn this qualification, however CP2 Modelling and CP3 Communications are not nearly as specialised now as they seemed when they were first introduced). I would also move the new core economics module recommended by the IFoA Economics Review Group I chaired in 2022 to replace the current CB2 module into the core practice section, as it would be more of a critical thinking module about the economic ideas necessary to underpin good actuarial work than the technical how-to subjects in core principles.
The specialist stage of the SP and SA subjects and the whole continuing education and additional certification options I would confidently expect the profession to retain, review and adapt as needed to reflect new challenges in members’ working lives. However, if CP1 is all that is needed post graduation to get to Chartered Actuary and CP1, two SP subjects and a SA subject gets you to Fellow, then suddenly time to qualification becomes much more predictable for students and their employers alike.
The risks to employers of taking on students would therefore be reduced, but also the rewards to taking them on will be much more transparent. The students we will be developing will have collapsed the gap between their credentials and their actual actionable skills, knowledge and experience. They will have:
Great team working skills;
Very highly developed presentation skills, both in writing and speech;
Strong IT skills and comfort working with data; and
Clarity about why they are in an organisation and a drive to use their skills to solve problems.
Epistemic rigour. They will be more likely to spot when a system or model is over-confident given the evidence, after an intense experience of interrogating models and evaluating evidence in their courses.
Synthesis. They will be able to integrate different perspectives into an overall understanding, through dissertation development, defending a position, understanding weaknesses in a position and adjusting accordingly.
Judgement. They will have been asked to make many judgements in their work and had to defend them in discussions with their supervisors and fellow students. Unconvincing opinions, not backed by evidence, will not pass muster.
Cognitive sovereignty. Students will have to take their own stand on their work, after all of the challenge and argument. Independence of thought is the hard-earned outcome here.
And then they will have a fighting chance of being centaurs in the world awaiting them, masters of the technology and opportunities available, rather than the reverse-centaurs that our education system is currently, and inadvertently, preparing them to be.
It all started for me in May 2017, with the challenge from Daniel and Richard Susskind in their book “The Future of the Professions”, which set out two possible futures for the professions. Either:
They carry on much as they have since the mid 19th century, but with the use of technology to streamline and optimise the way they work; or
Increasingly capable machines will displace the work of current professionals.
Their research suggested that, while these two futures would exist in parallel for some time, in the long run the second future would dominate. The actuarial profession was going to be particularly vulnerable. As the Susskinds wrote:
Accountants and consultants, for example, are particularly effective at encroaching on the business of lawyers and actuaries.
I stood in the Institute and Faculty of Actuaries Council elections that month, on a platform saying that we needed to urgently respond to this challenge. I didn’t get elected.
However, at the University of Leicester we pursued a curriculum transformation programme in response to this challenge aimed at developing actuaries of the future who had:
Highly developed presentation skills, both in writing and in speech
Great team working skills
Strong IT skills – comfortable with working with data
Clarity about why they are there and the desire to use their skills to solve problems
The 2017 post also talked about emerging trends which had hardly started at all yet:
The end of reserved roles for actuaries
Different ways of communicating advice
Online self-help for users of actuarial advice
The advance of roboactuaries and their assistants
A paper called 2036: An actuarial odyssey with AI written for the Society of Actuaries in July 2016 by Dodzi Attimu and Bryon Robidoux discussed the possibilities for robo actuaries and robo actuarial analysts.
They estimated that robo actuarial analysts would be with us in 5-10 years and would provide:
A system that has limited cognitive abilities but can undertake specialized activities, e.g. perform the heavy lifting in model building (once the specification/configuration is created), perform portfolio optimization, generate reports including narratives (e.g. memos) based on data analysis, etc.
Whereas a robo actuary was more like 15-20 years away and which they also helpfully described:
We mean a software system that can more or less autonomously perform the following activities: develop products, set assumptions, build models based on product and general risk specifications, develop and recommend investment and hedging strategies, generate memos to senior management, etc.
Then, in 2018, there was the whole Bullshit Jobs argument posed by the late David Graeber, discussing Keynes’ prediction in 1930 that:
In quite a few years – in our own lifetimes I mean – we may be able to perform all the operations of agriculture, mining, and manufacture with a quarter of the human effort to which we have been accustomed.
Graeber, in Bullshit Jobs, pointed out that this never happened, despite pretty much all of the technological developments and income increases which Keynes predicted. He suggested that the future which the Susskinds were predicting is already happening in terms of needing fewer people to fill the meaningful roles within organisations but that, rather than employing fewer people, we are either creating “bullshit” jobs which even the people doing them can see no point to or bullshitizing existing roles for which the meaningful need has passed. It was as if the organisations themselves have attempted to maintain the outward appearance of the same structures by disguising the hollowing out of so many of their functions with simulated business.
Why? One of the reasons he thought the situation had been allowed to develop was that noone believed that capitalism could produce such an outcome. Graeber gave the example of the creation of Obamacare, where Barack Obama “bucked the preferences of the electorate and insisted on maintaining a private, for-profit health insurance system in America” in order to protect jobs in health insurance.
Then we had the pandemic, and the painful return to work which found that people were not necessarily keen to return to populating those office empires, preferring to work remotely. Some of the attempts by the captains of industry to get them back were a little desperate.
Now nine or ten years on from the initial challenge, we are deluged in articles about how AI is impacting different areas of actuarial work, whether it is already replacing graduate roles and what actuarial students need to do to make themselves employable. And now the blinkers also seem to have come off about capitalism not producing the need for fewer jobs.
Historically, accountancy firms have typically had a pyramid structure – wide base, heavy graduate recruitment. Firms are now starting to talk about a ‘diamond model’ with a wide middle tier of management because, ultimately, AI is not sophisticated enough yet to make those judgment calls.
But hang on a moment. Now there is something called an “AI boomerang”. Sam Altman of Open AI and Dario Amodei of Anthropic have both been backtracking on their predictions of job lay offs due to AI. As the Gold and Geopolitics Substack puts it:
Two-thirds of the companies that ran AI-driven layoffs last year are now rehiring. One in three spent more on the rehiring than they saved on the original cuts. Robert Half calls them “AI boomerangs” – which is the name a consultant invents when there’s a market in unwinding what the last consultant sold you.
The last few years of AI-generated headlines (in both senses of the phrase) have been quite the rollercoaster.
I do not want to add to the deluge, as currently you would need a LLM summarising reports for you 24/7 just to keep up with it as it is.
And there are good reasons for not rushing to judgement on this. It is only six weeks ago that the Bank of England’s Head of Financial Stability warned that global stock markets are overvalued. As she said:
The thing that really keeps me awake at night is the likelihood of a number of risks crystallising at the same time – a major macroeconomic shock, confidence in private credit goes, AI and other risky valuations readjust – what happens in that environment and are we prepared for it?
Since then, the economics of the major AI players has only got more bonkers. In this environment, there is considerable uncertainty about what students should be learning to prepare them for the world they will need to rebuild from the rubble of the current one.
Daniel Susskind reckons a “no regret” strategy for education would focus on the basics, which he describes as literacy, numeracy and critical thinking, and the critical use of AI. Carlo Iacono talks about the work which will always be hard to automate as follows:
The work that depends on reading a room. The work that relies on institutional memory. The work where the important fact is not in the document. The work where the answer is political, ethical or relational as much as technical. The work where being right is not enough, because someone has to be accountable for the consequences.
But you have to get into the organisation first before you can develop much of this. I remain reasonably comfortable with our prescription from 2017 in terms of broadly what the curriculum should be trying to achieve. Curriculum changes take time and, tempting though it is to pile into syllabus changes aimed at incorporating the latest cutting edge technological developments, the likelihood is that you will be arriving at the wrong fire. As a social media post this week put it:
The danger is that, as Agentic AI in particular looks so much like the predictions of 2016, we declare it the new messiah and bet everything on this being the future. I certainly think that 2036 will look very different to now, but I am not convinced that we have the shape of it yet. This may just be another bullshit alternative, waiting for the next crisis to mutate into something else.
What we can say with certainty is that the future for early professionals is as uncertain as any of us can remember. And, in my view, the best way to support our young people starting their professional lives (or whatever is going to replace professional lives) is to make the value they add much clearer to anyone who might want to work with them in the future.
Future education systems are going to need to help their graduates demonstrate their value in ways they haven’t historically needed to. AI has not brought new problems, it has accelerated existing ones. And the gap between actual actionable skills, knowledge and experience and the credentials which are supposed to represent them is the key one I believe as far as those graduates are concerned.
Because currently the companies who might want to take them on have to guess what they can do to a large extent. What does a 1st mean? A 2.1? Do the differences matter to their employer? Should they interrogate the details of student courses? How would they incorporate that into a manageable recruitment process? Some people see a future of AI generated applications by the thousand per student doing battle with AI powered triage systems operated by potential employers.
This is already an issue in academia. The conclusions of one academic paper on submissions to one major academic journal were stark:
Submission volume has risen 42% since the late 2022 release of ChatGPT, while writing quality has declined. The rise in AI-generated writing accounts for nearly all of these trends.
On the other hand, researchers complain about the increasing use of AI peer review to try and cope with these increased volumes, and some have tried to game the LLMs they believe they are dealing with.
I think we can do better than this in actuarial teaching and learning.
I didn’t understand statistics until I started taking actuarial exams that required me to master particular statistical techniques, decide which ones were appropriate to the problem I was looking at, apply them accordingly and be able to interpret what the results did and didn’t tell me. An A at GCE O level in maths, and A at GCE A level in both maths and further maths and a maths degree from Oxford did not give me those abilities.
I didn’t understand economics until I started putting together modules which could be taught on both BSc and MSc courses and then teaching them. My economics module on the way to qualifying as an actuary, based very heavily on Economics by Begg, Fischer and Dornbusch, did not give me that understanding.
There is a pattern here: the most impactful experiences we have are frequently at a bit of a distance from the credentials we present to the world. Your education is not a paragraph on your CV, it is your lived experience, sometimes assisted by, sometimes actively hindered by and often pursued completely independently of the educational institutions you have had a relationship with during your life.
Daniel Susskind has marched into this often fraught relationship between credentials and actual actionable skills, knowledge and experience in the last lecture of his Gresham College series on The Future of Work, called Education – And Its Limits. His analysis is a very clear expression of the problems that will be created if AI systems prove to be half as capable and long-lasting as people from OpenAI and Anthropic are telling us they will be.
Susskind argues that trying to future-proof students through education was a hopeless task and that working on the assumption of unresolvable uncertainty was the better way forward. He suggests a “no regret” strategy for education would focus on the basics, which he describes as literacy, numeracy and critical thinking, and the critical use of AI. Others in the audience suggested some other “basics”: communication skills for instance.
The other strand of Susskind’s basics was critical use of AI. And his challenge to the audience was whether we can teach AI without losing the basics in the process.
After a bit about the need to make continuing education in later life more accessible, he looked at his list from a previous lecture (which I briefly touched on here) of problems for a post AI future:
Distribution (replacing wages);
Contribution (how do you “pull your weight”);
Power (domination by Big Tech on economics, politics, liberty, social justice and democracy); and
Meaning (fulfilment in life).
And this is why I continue to watch Susskind’s output, because he is the unusual combination of an extremely orthodox economist (see his book Growth: A Reckoning for proof of this) and someone who has been wrestling with the challenges of more capable systems to our way of doing things for over 10 years. What Susskind gives you is a peek at how the actual economists advising our governments would deal with things if OpenAI and Anthropic are proved right. It is as if you had someone who both thought, as Ben Bernanke, Chair of the Federal Reserve, did in October 2007, that “the banking system is healthy” and also that the banking collapse was going to happen anyway.
Because what he reveals is that, if Anthropic are right, orthodox economists have really got no policy prescriptions worthy of the name.
On 1. Distribution: Susskind acknowledges that, if the labour market could not redistribute wealth effectively any more, then a bigger state would be needed to do so (but he was at pains to emphasise that this would not be the 20th century central planning type of state).
On 2. Contribution: Susskind thinks that perhaps we should allow people to make non-economic contributions! As if all of the activity in society which economists routinely ignore really wasn’t already happening!
On 3. Power: Susskind says the political power of Big Tech with regard to liberty, social justice and democracy is a problem. We have anti-trust legislation that can deal with Big Tech’s economic power, but not its political power. I don’t know what kind of political power he thinks Big Tech would have without the economic power that has been granted it by a steady erosion of that anti-trust legislation over recent years.
On 4. Meaning: he has little to say other than something about us currently having policies for work but not for leisure.
And in all of this, there is the implicit underlying assumption of a stable future environment for all of this tech to operate within.
For me it brought to mind something a friend of mine who was in a tent with Cory Doctorow at the How The Light Gets In festival at Hay-on-Wye last week. “He has a theory of change” he said.
He really does, including about how to break the economic power of the Big Tech companies. You all need to read Enshittification for the full account, but my review of the book here is a sneaky peek.
Susskind really really does not have a theory of change. Which tells me that the economics profession does not have one either.
However I continue to watch him as I find he goads me into thinking what some better answers might be to the questions he asks. And perhaps also a better question than whether we can teach AI without losing the basics in the process. I think a better question would be what do we need to learn when the future is uncertain.
First of all, let’s remind ourselves of the problem. If noone cares how your advice was constructed, but your client can get advice that ticks the compliance box more cheaply and quickly from an AI system, while the experienced professional still has some role in managing the process, it may increasingly be a struggle to justify the cost of the junior colleague. So the future education system is going to need to help that future junior colleague demonstrate their value in ways they haven’t historically needed to. AI has not brought new problems, it has accelerated existing ones. And the gap between actual actionable skills, knowledge and experience and the credentials which are supposed to represent them is currently the key one as far as that future junior colleague is concerned.
The temptation is to rush into syllabus changes towards what currently looks like the cutting edge activity. I agree with Susskind here that this would be a mistake. He cites the example of Michael Gove, amongst many other education ministers at the time, mandating the teaching of coding in 2014. Now we find that the new AI systems (despite the problems highlighted by Hannah Fry, Kyle Kingsbury and others I talked about here) are most suited to writing code and Anthropicclaim that Claude is now writing 80% of its own code. Programmers are saying that, on the famous XKCD cartoon above, they are now living on the theory curve.
But both higher education institutions and the professions who still want to be in the game of developing the next generation of professionals can do a lot more both to reduce the gap between credentials and actual actionable skills, knowledge and experience and to make it clearer to employers that they have done so. That will be the subject of my next post.
Source: Nick Foster – December 2013 – originally drawn to deride George Osborne’s austerity as Pugwash Economics, repurposed now as I am worried about the actuarial profession pulling up the ladder on the next generation
The “black box” was a constant refrain when I was working as an actuarial consultant. It was where the results from a process were being accepted without any understanding of how they were arrived at. Something we felt that any self-respecting actuarial consultant should challenge in their own work and everybody else’s.
However when you came to actually present analysis or arguments to a client, you expected a certain amount of that expertise to be taken as read, to effectively be inside a black box as far as the client was concerned. They couldn’t be expected to understand all of the aspects of what you were talking about, otherwise they wouldn’t need you. Good practice was always to put them in a position where they could understand and make decisions about the key aspects of your advice without needing to engage with the other parts. As the expert, you decided what was in the black box.
Now the black box is back with a vengeance for all the professionals who have relied upon them in their working lives. As Dan Davies puts it:
The same black-box property which stops you from being second guessed or overruled means that nobody is interested in your explanations for your decisions; it is definitional of being a black box that you are going to be judged by results.
And, if you are in the business of advising in the teeth of uncertainty, as actuaries are, then this is likely to be a real problem. If noone cares how your advice was constructed, but they can get advice that ticks the compliance box your client has to complete more cheaply and quickly than you can, the more automated black box is going to win the business. The experienced professional still has a role in managing this process, verifying the results coming out of the black box and determining what can still be kept out of the black box, but he may be increasingly struggling to justify the cost of his junior colleague.
Well things don’t look so bad in the UK right now according to the Office for National Statistics (ONS), reverting to close to the average after a post pandemic surge in the finance and insurance sector:
However, if we look at the United States, which tends to show us where the UK finance sector is going, it looks far more ominous:
Yesterday Sky News ran a story about Standard Chartered‘s CEO who, in his desperation not to describe over 7,500 job losses as cost cutting, said this:
It’s not cost-cutting. It’s replacing in some cases lower-value human capital with the financial capital and the investment capital we’re putting in.
We may need to sit with that statement for a little while.
Daniel Susskind talks about this risk in his latest lecture entitled A World Without Work: in summary, to paraphrase only slightly, sure relatively junior white collar roles may already be particularly hard hit by AI, but he is optimistic because of the impact on GDP and we cannot pause because of China. He then goes on to talk about the four problems he sees for a post AI future:
Distribution (replacing wages);
Contribution (how do you “pull your weight”);
Power (domination by Big Tech on economics, politics, liberty, social justice and democracy); and
Meaning (fulfilment in life).
Susskind has gone from thinking that the fear that AI is coming for your jobs was overblown and that it was just task encroachment that we faced, to now thinking that it may encroach on all the tasks in most fields. Jevons Paradox (that technological innovation that increases the efficiency of a resource’s use leads to a rise in consumption of that resource) is no comfort if that new demand is robot-met.
Carlo Iacono suggests that the move of junior roles to AI may be subtle to begin with:
The weakness among young workers may appear as fewer people entering employment from outside the workforce. Firms may not fire large numbers of juniors; they may simply hire fewer of them.
That matters. The labour market can look healthy while the entry path narrows. Senior workers stay employed. Output rises. Productivity improves. There is no dramatic wave of redundancies.
Yet the first rung is being taken out.
It may also be masked by the fact that there remains a shortage for actuaries beyond the entry roles. There is almost a hint of desperation to approaches like this looking for introductions from a retired actuary like myself:
(followed by a list of clients he is working for)
Source: recruitment consultant who will remain anonymous. I am assuming “candies” are candidates
And, even if you think the risk of the AI Bubble bursting soon, taking down the global stock markets underpinned by the Magnificent 7, is exaggerated, you do need to be suspicious about the current abilities of AI to replace junior staff. My experience with another, somewhat earlier, actuarial technology, the pensions valuation engine, would suggest that the outputs need to be analysed very carefully before sharing with a client: it often had dependencies between what should have been independent variables hidden in the programming, or vagaries in the setup which left out non-standard benefit rules for your particular scheme, for instance. Or the student who had set it up initially (a complicated process usually) might have made a mistake or you might have not communicated with them very well to start with. Or a hundred other things.
For whatever reason, there was often still a lot to do after the valuation engine had produced some output.
Can this sort of thing happen with agentic AI? Well think about that student programming the valuation engine, but on steroids. Its patchy capabilities combined with its basic psychopathy leads to, as Hannah Fry entertainingly demonstrates here, some serious problems arising with the agent’s relentless to and fro with the large language models it depends upon, asking them what it should do next. As Hannah says:
I built an AI agent. She opened a shop selling novelty mugs, emailed a journalist without being asked, and then leaked our passwords to a total stranger.
As Kyle Kingsbury wrote about having an AI agent as a colleague in a programming team:
Imagine a co-worker who generated reams of code with security hazards, forcing you to review every line with a fine-toothed comb. One who enthusiastically agreed with your suggestions, then did the exact opposite. A colleague who sabotaged your work, deleted your home directory, and then issued a detailed, polite apology for it. One who promised over and over again that they had delivered key objectives when they had, in fact, done nothing useful. An intern who cheerfully agreed to run the tests before committing, then kept committing failing garbage anyway. A senior engineer who quietly deleted the test suite, then happily reported that all tests passed.
You would fire these people, right?
Yet despite all this, the money continues to pour in to the construction of AI infrastructure. There are already websites up and running for all of the parts of tasks AI cannot encroach upon.
The bottom rung of the actuarial ladder is clearly in danger. This is a particular problem for the actuarial profession, which has traditionally relied on longer periods of work-based training for its future qualified actuaries than many other professions. Training to become an actuary takes a long time. Median time to fellowship is still around six years, with some taking up to ten or giving up. The exams are hard to pass. There have been attempts by the profession to tackle some of these disincentives: the Chartered Actuary designation to make a destination of the generalist qualification before the specialisation of the fellowship, championed on this blog and launched in the teeth of opposition by some fellows, being one example.
It has led to a culture within actuarial firms around managing the extended time in training, with rituals around study leave and results days. One of the fears expressed in opposition to the introduction of the Chartered Actuary designation was that, if this could be achieved almost entirely within formal education at universities, the value of working alongside experienced actuaries would be lost.
It has led to a culture within the profession itself of managing large parts of its education system in house. Half of its revenue and around 30% of its expenditure are on “pre-qualification learning and development”. Sometimes it looks more like an education business with a professional side hustle.
But then the new AI toys have come along, and it turns out that many of those experienced actuaries may be less keen on graduates coming in and needing supervision from them after all. Many of them may rather spend hours on AI prompts than on developing another human being.
I fear that, increasingly, companies are not going to accommodate actuarial students in their work plans without significant persuasion. And, if the number of students studying while in work falls, the profession itself is going to struggle to finance its own bespoke education system at an acceptable cost to its members.
It will be hard for the profession to challenge this too: it is going to be good for many of those already established in their roles as the market for more experienced actuaries, when the market has no interest in developing the actuaries of the future, becomes increasingly competitive.
If the actuarial profession does accept the challenge of protecting the pipeline of future experienced actuaries it will need to review its entire education syllabus through this lens. It will also need to engage with other partners involved in what is in effect a problem of capital formation and collective action: government incentives may be needed to encourage firms to continue to train early career professionals and discourage free-riding. There may be no way back for the student with no actuarial qualifications learning on the job. The universities may be needed to plug people in at a different career point, which will require them to innovate themselves even further into the professional training role than ever before. As Carlo Iacono points out:
educational institutions may be pushed to simulate more of the apprenticeship environment. That does not mean adding a thin “AI literacy” module. It means creating settings where students practise judgement under uncertainty, in realistic workflows, with feedback that is close enough to hurt and useful enough to teach.
It will not be at all easy. But the alternative is a future without opportunity for those who do not already have it and an ageing profession withering on the vine it refused to nurture.
I have caught Covid for the third time this week, so naturally my thoughts have turned to how it all began.
There are a few Covid posts starting to turn up online as the 6th anniversary of it all rumbles around. The British Foreign Policy Group have helpfully published a timeline from which I have taken everything that happened before Boris Johnson locked us down for the first time:
So a lot had happened by 23 March. You will all have your favourite bits from the saga above, I think mine is 22 January, when Public Health England announced they had moved the risk level to the general public from very low to low.
I remember teaching a macroeconomics class on 12 March when we knew it was going to be the last session on campus. The penny hadn’t dropped. Students were asking about how they would hand work in. We agreed it would have to be online. Some lecturers were talking about microwaving paper submissions to sterilise them. We had a little giggle about that. I had spoken to Stuart McDonald (now MBE) earlier that day where we had reluctantly agreed to postpone his visit to campus to speak to the Leicester Actuarial Science Society (LASS). Stuart would of course become one of the actuarial stars of the pandemic for his work with the COVID-19 Actuaries Response Group. I had a similar conversation by email with Lord Willetts, who was Chancellor at the University of Leicester at the time and who was going to talk to LASS about his books The Pinch and A University Education. We talked of postponing rather than cancelling. The realisation that everything was changing for the foreseeable future was still not there.
It took a long time for the penny to drop for the Government as well. As this analysis of the establishment of the “Covid Disinformation Ecosystem” says:
January featured fear and disbelief, February proved covid couldn’t simply be ignored, March was when governments realised the hospitalisation rate could overwhelm healthcare.
And a Government that was slow to respond initially was very vulnerable to the groups which sprung up during 2020 and 2021. As the Counter Disinformation Project says:
And the main initial target for the UK section of the ecosystem was Boris Johnson who was meeting privately with newspaper owners and editors. Enough doubt was put into Johnson’s mind that he dithered and delayed when cases began to rise, leading to a private meeting with Heneghan, Gupta and Sweden’s Anders Tegnell in September before he chose to ignore his scientific advisors’ calls for a circuit breaker lockdown. In the run up to the deadliest weeks of the pandemic the papers were calling for Johnson to “Save Christmas’.
However I don’t want to focus on our collective inability to make decisions during crises this time. This time I want to focus on the impact of the pandemic on our mental health.
By coincidence, today the 386 page Module 3 report from the Covid Inquiry on The impact of the Covid-19 pandemic on the healthcare systems of the United Kingdom was published. The longer this Inquiry goes on, the more it appears to resemble a truth and reconciliation commission rather than something likely to improve the handling of future pandemics. It gets past transgressions on the record, but in a way designed to move us on rather than improve our preparedness and organisation. I certainly saw nothing in the summaries that I didn’t already know. Module 3 has made 10 recommendations. The only one which mentions mental health at all is the last one on Psychological and emotional support for healthcare workers.
Looking through the module titles, it would seem that this is unlikely to be rectified until Module 10 – Impact on society – reports, currently scheduled for the first half of 2027. I find this relegation of our collective trauma to the lowest priority astonishing.
Data on the prevalence of mental health difficulties is harder to assess. For children and young people, surveys in England have provided a time series since 2020 that suggests very strongly that mental ill health is indeed more prevalent now than it was before the start of the pandemic. A steady rise in the decade prior to 2020 seems to have been followed by a sharp rise, and numbers have stayed high ever since. We do not have the equivalent data for adults, meaning that a clear picture has yet to emerge, but there is persuasive evidence that levels of mental ill health have been rising over the last decade, and the pandemic has contributed to many of the risk factors people face.
Before concluding as follows:
Crucially, the pandemic exposed fault-lines in the nation’s mental health, and the stark inequalities faced every day by people living with mental illness. The public’s mental health was deteriorating in the years running up to the pandemic, and mental health services were struggling to deal with the consequences of many years of underfunding and austerity measures across public services. People with a mental illness were already dying 15-20 years sooner than the general population, and facing widespread hardship. The pandemic exacerbated these inequalities, creating new risks to people’s mental health and reducing access to support.
We now have the opportunity to learn from this experience and build a mentally healthier future. We can act now to boost the public’s mental health in the aftermath of the pandemic, protecting those who have experienced the worst effects and offering better support to groups that don’t yet have access to the right support. And we can incorporate mental health into preparations for future emergencies, so that responses are psychologically informed from day one.
They also made 10 recommendations, mostly for the NHS and Department for Health and Care, but also covering education, communications and considerations for the upcoming (at the time) review of the Mental Health Act. Less than half of these recommendations have been addressed at all.
Now we are two years on from that report, what has changed?
Well, Roy Lilley has drawn a rather dispiriting picture for us. He draws attention to Wes Streeting’s announcement in the Health Service Journal on 12 March, that the proportion of the NHS budget spent on mental healthcare would be cut for the third year in a row. Lilley lists how the demands on mental health services have mushroomed since before the pandemic:
Around two million people were in touch with mental health services in 2019, today it’s around three million;
Child and Adolescent Services: in 2019 around 500,000 referrals. Now around a million;
And only around 45% of referrals are accepted, meaning the true demand is even higher;
Talking therapies are up by 60%; and
Crisis team referrals and sectioning under the Mental Health Act are also up 60%.
And he summarises the problem like this:
The total economic cost of mental ill-health in England in 2022 was estimated ~£300bn a year when lost productivity, welfare and wider costs are factored in.
The total MH budget is about £16bn. Meaning, the NHS is spending roughly £1 trying to address a £18 national problem.
It feels like we are still waiting for the penny to drop.
A week or so ago I referred to a “Thought Exercise” set in June 2028 “detailing the progression and fallout of the Global Intelligence Crisis” (ie science fiction), published on 23 February, which may have tanked the share price of IBM later that day. As I said then, the fall definitely happened, with IBM’s share price falling 13%, its biggest fall since 2000. I said then that the likelihood of the scenario portrayed was difficult to assess, but the speed with which the total economic collapse was described felt unlikely if not impossible. I would like to expand on that.
The main reason that the scenario was hard to assess was that it was not based on data or evidence at all. That is unavoidable for speculative fiction talking about things that are not currently happening, but when describing an economy only two years away where most of the processes described should be discernible to some extent already, it is totally avoidable.
Ed Zitron has done an excellent line by line take down of the Citrini piece here. Here is one page of that to give you a flavour:
However this lack of a link with anything tangible did not stop the financial markets panicking, which should cause us pause when relying on the financial markets’ valuation of projects, industries, government policies, etc.
Ed Zitron describes this kind of piece as analyslop: “when somebody writes a long, specious piece of writing with few facts or actual statements with the intention of it being read as thorough analysis”. It can then get picked up by other commentators which take it as their starting point for further analysis, often making it hard to see that the starting point had few if any data points. Here is an example, from Carlo Iacono, looking at what if just some of the Citrini pronouncements were true, with appendices detailing possible branching paths of outcomes, all generated by a large language model (LLM). And then people start studying the meta analysis, and it starts getting taken even more seriously, and put into models and pretty soon most of the analysis is being done on imagined risks rather than on ones which are already staring us in the face.
We have always had a problem keeping our society grounded in reality, think the 2003 Iraq War, where we went to war on a false assessment about Iraq’s possession of weapons of mass destruction, the 2008 financial crisis, where banks misunderstood the risks they were exposed to, and the last two and a half years, where we, for the most part, seem to have convinced ourselves we have not been facilitating a genocide in Gaza when we clearly have been. But this is only going to get worse with the AI systems which are being developed.
The rapid rise of artificial intelligence has served to dramatically increase the speed of information production while also eroding accuracy, making it difficult to differentiate between content that simply sounds confident and content that’s actually grounded in reality.
So where is AI currently? Well PwC’s global CEO survey from January this year had the following statement as the first bullet amongst its key findings:
Most CEOs say their companies aren’t yet seeing a financial return from investments in AI. Although close to a third (30%) report increased revenue from AI in the last 12 months and a quarter (26%) are seeing lower costs, more than half (56%) say they’ve realised neither revenue nor cost benefits.
That’s the reality. But the hype is much much more entertaining. My favourite spoof video of the AI future currently is this one, about the time where all most of us are good for is riding bicycles to supply the ever increasing energy needs of AI systems (click view in browser if you can’t see it):
And what about the financial journalists? The pieces describing our reaction to whatever is about to unfold economically have already been written. There are investor websites asking if the 2026 crash has already begun, while another recent article argues that “America has quietly become one of the world’s most shock‑resistant economies” (which seems unlikely to age well). What most financial journalists are more comfortable with are articles about how the warnings were ignored after the fact.
And the professions? Well the current overview of my own profession is probably reasonably represented by this piece from the Society of Actuaries in the United States. Unfortunately for them, Daniel Susskind, who is mentioned in the article, is currently suggesting, as part of his Future of Work lecture series for Gresham College, how the key to the sudden development in AI, after the “AI Winter” when progress seemed slow, was that we abandoned trying to make machines which thought and acted like humans in favour of focusing on completing tasks in any way possible. Increasingly we are now automating tasks where we can’t (or won’t) articulate how we do them. From Deep Blue‘s victory over Kasparov in 1997 to Watson winning jeopardy in 2011 to ImageNet beating humans at image recognition (although that is disputed), Susskind refers to this progress as the displacement of purists in favour of what he calls “The Pragmatic Revolution”. Pragmatism in this sense appears to be that we humans should just accept the consequences the people running these systems want. So, as his latest lecture “Work, out of reach” claims, people moving into cities to find work is a strategy which is no longer going to work for low skilled people:
He then shows this graphic demonstrating the lack of recovery of big coal mining areas in the UK:
Source: Left – Sheffield Hallam University map of coal mining areas; Right – % employment from Overman and Xu (2022)
And finally he cites the notorious Policy Exchange piece from 2007, Cities Unlimited, whose thesis was that there is apparently no realistic prospect of regenerating towns and cities outside London and the South East.
Susskind talks about three forms of technological unemployment:
skills-mismatch, where your skills are mismatched to the work available. Education and training has always been the answer to this in the past.
place-mismatch, where the jobs are not where you have built your life. Some believe the answer should always be the one proposed by Norman Tebbit, who memorably told everyone in 1981, “I grew up in the 30s with an unemployed father. He did not riot. He got on his bike and looked for work.”
identity-mismatch, where according to Susskind, people are prepared to stay out of work to protect their identity, citing US men who won’t take “pink collar” work, China “rotten tail” kids, Japanese seishain-or-nothing and Indian Sarkari Naukri queues in India. Or perhaps they are just looking for work which is consistent with the idea of human dignity.
Susskind claims to have no answer to any of these as far as AI is concerned. They are, in his view, just the inevitable outcomes of his “Pragmatic Revolution”. It is the unthinking pursuit of more and more growth funded by capital less and less tethered to any territory, principle or purpose, where any grit in the machinery, be it unions or protestors or, increasingly, the wrong sort of government must be trampled underfoot. Anything which impedes the helter-skelter rush to more and more at greater and greater speed. It’s like our whole economy is run by this guy (press the view in browser link if you can’t see him) shouting “Ready, Aim, Fire!”:
But unskilled people will not be the only collateral damage of these unguided weapons. Take markets for instance. These are where people are exposed to risks and rewards based on underlying conditions they only partially understand. Greed and fear may be their main motivations, but gossip and group think are their main communication channels. They don’t need facts, particularly when so many of the facts are proprietary information not in the public domain. A plausible narrative will do. And plausible narratives are what LLMs will do for you in abundance.
And the more we reward people who can move fast, eg to spot an arbitrage opportunity, even at the risk of breaking things, rather than people who can make decisions which still look good decades from now, the more we are setting up the conditions for AI systems to be the go-to tool.
And put that together with an AI industry which desperately needs funding capital to keep arriving, ie one which is unbelievably highly motivated to push plausible narratives even when they know they are not grounded in reality, and you have a recipe for market-generated chaos.
And then we have Trump’s new war. Beware the people who are war gaming the Middle East at the moment on a range of LLMs (just stop and think for a moment about the bloodless inhuman impulse behind carrying out such an exercise rather than, I don’t know, talking to some actual people who live or have lived recently in and around the region). One of the worst offenders is Heavy Lifting banging on about what the three scenarios are for Operation Epic Fury. This is as bad as it sounds:
I tasked her [he is talking about Gemini Pro here] with doing a literature review on regime change (a term often used by the President but not a well-defined one), creating three scenarios of possible outcomes for which each was give a percentage probability, and a list of 20 items to examine for each scenario that covered political, economic, and cultural issues with a special focus on the political consequences in the U.S. and what this means for China, our biggest geopolitical rival.
But Gemini Pro wasn’t the only one involved in this. Two other humans were, Tim Parker and Ron Portante, trainers at the gym I go to. (Just as a personal aside, Tim was my coach in hitting six plates [345 pounds] on the sled last Friday and I have a video to prove it!) I was talking about the piece and Ron raised the issue of linguistic and cultural diversity in Iran. Tim did some real time research for me on his phone while I was burning real calories under his strict tutelage. This made me think I needed a background section on Iran. When I got him Gemini and I added it.
What you mean you belatedly realised you might need to have done some actual research into Iran rather than just generic research on regime change? I stopped reading at that point.
Meanwhile King’s College London have been carrying out war games more systematically using AI. Professor Kenneth Payne from the Department of Defence Studies led the study, which looked at how LLMs would perform in simulated nuclear crises. As Professor Payne said:
Nuclear escalation was near-universal: 95% of games saw tactical nuclear use and 76% reached strategic nuclear threats. Claude and Gemini especially treated nuclear weapons as legitimate strategic options, not moral thresholds, typically discussing nuclear use in purely instrumental terms. GPT-5.2 was a partial exception, limiting strikes to military targets, avoiding population centers, or framing escalation as “controlled” and “one-time.” This suggests some internalised norm against unrestricted nuclear war, even if not the visceral taboo that has held among human decision-makers since 1945.
This is not a Pragmatic Revolution. These AI systems cannot replace humans thinking about the future we want for humans in any way which is worth having. What they can do, if we let them, is accelerate our worst impulses and move us further away from considered reflective decision making.
But we will continue to use AI systems in the military because, as it turns out, it is very useful for low stakes admin. So although Lavender, the system used by the Israeli military to select targets in Gaza, made errors in 10% of cases and was therefore totally inappropriate to the task, there are lots of organisational logistical tasks where it is much quicker than the alternative and 10% error rates do not matter so much.
There is clearly an issue with what we decide to use these systems for. We need to be able to regulate the decisions which are particularly consequential. However the only way we seem to be considering for this at the moment is the human-in-the-loop model, like the humans spending around 20 seconds considering each target recommended by Lavender before authorizing a bombing. I have written about these before in the context of early career professionals in the finance industry, where the prospect seemed miserable enough:
They will be paid a lot more. However, as Cory Doctorow describes here, the misery of being the human in the loop for an AI system designed to produce output where errors are hard to spot and therefore to stop (Doctorow calls them, “reverse centaurs”, ie humans have become the horse part) includes being the ready made scapegoat (or “moral crumple zone” or “accountability sink“) for when they are inevitably used to overreach what they are programmed for and produce something terrible.
However it seems obvious to me that, in the context of dropping actual bombs on actual people, there is an even more serious problem with this model. As Simon Pearson (anti-capitalist musings) puts it:
The “human in the loop” requirement exists in military doctrine because international humanitarian law demands an accountable human decision-maker for lethal force. The laws of armed conflict require proportionality assessments, precautionary measures, distinction between combatants and civilians. All of these obligations attach to a human commander. The system cannot fulfil them. So a human must be present, and their presence must constitute a decision, regardless of whether any genuine decision was made.
What the institution needs from the analyst is not judgment. It is a signature. The signature converts a machine output into a human act. And a human act is what the law recognises, whether or not any judgment occurred. When the strike kills children, the chain of accountability runs to the analyst who approved the target: not to the system that identified it, not to the company that built the system, not to the doctrine that compressed the review window to ten seconds.
But whether we want to make money from exploiting a short term anomaly in a market, make our fellow humans redundant, prosecute a war on another group of fellow humans or “win” a war of mutual nuclear destruction, we need to retain the capacity for real human reflection within the decision-making processes we use. Not just a human-in-the-loop nor just the elites of tech companies deciding how the systems will be configured behind commercially confidential walls. These processes need democratic accountability every bit as much as our parliaments, councils, institutions and voting systems do.
Something infuriatingly slow, inclusive and deliberative giving recommendations which are then stress-tested for how they would perform on contact with reality, involving yet more people being serious and deliberative and taking their responsibilties more seriously than being a human-in-the-loop would ever allow. Our decision-making systems need more grit and less oil. AI is all oil.
The Actuary magazine recently had a debate about whether the underlying data or the story you wove around it was more important. I’m not sure there is always a clear distinction between the two, as Dan Davies rather neatly illustrates here, but my view is that, if a binary choice has to be made, it is always going to be the story. And there was a great example of this which popped up recently in the FT.
The FT article was ‘Is university still worth it?’ is the wrong question, by John Burn-Murdoch, with great graphs as usual by John. However, as is sometimes the case, I feel that a very different and more convincing story could be wrapped around the same datasets he is showing us.
The article’s thesis is as follows:
The graduate earnings premium, ie how much more on average graduates earn than non-graduates, has only fallen in the UK as the proportion going to university has risen. It has risen in other countries:
In the UK, we have had much weaker productivity growth than the other comparator countries, and also “the steady ramping up of the minimum wage has squeezed the earnings premium from the lower end too”:
We have also had a much smaller increase in the percentage of managerial and professional jobs than a different group of comparator countries (they haven’t mentioned Germany before), meaning graduates are forced to take lower salaried jobs elsewhere:
So the answer according to the FT? We should focus on economic growth rather than “tweaking” higher education intake and funding. Then graduate earnings would be higher, student loans could be more generous(!) and students would have more chance of getting a good job.
Well perhaps. But here’s a different framing of the same data that I find more persuasive.
Let’s start by addressing that point about the minimum wage. According to the House of Commons Library report on this, the UK’s minimum wage is broadly comparable to that of France and the Netherlands, although higher than Canada’s and much higher than that of the United States. The employers who are the FT’s constituency would obviously like us lower down this particular chart:
The main economic framing here is the progress myth of the UK’s business community: economic growth. All problems can be solved if we can just get more economic growth. Apparently we need more inequality in pay between graduates and non-graduates which we can get by generating more economic growth. This is honest of them at least, although I don’t see much evidence that the economic growth they crave will go into skilled job creation rather than stock buy backs (according to Motley Fool, “Companies spent $249 billion on stock buybacks in Q3 2025, and $777 billion over the first three quarters of 2025.”).
There are a lot of problems with framing every economic question with respect to economic growth, memorably illustrated by Zack Polanski of the Green Party in this less than 3 minute video recently (I strongly recommend you watch it before you read on – click on the read in browser link if you can’t see it):
Economic growth is increasingly without purpose, wasteful of energy and poorly distributed. It is chasing outputs, literally any outputs, whatever the cost to the environment, our health system, our education system, our social support systems and our communities. Looking at the framing above, you can see that economic growth as currently pursued will always see anything which stops the concentration of wealth amongst the already wealthy, like a higher national minimum wage or a totally made-up concept like a lower graduate earnings premium (which in itself is a framing trying to make reducing inequality seem undesirable) as a problem. Lack of productivity growth, itself a proxy for this kind of economic growth (because if you ask why we need more productivity the answer is always to get more economic growth), is usually directed as a criticism at “lazy” UK workers, rather than under-investing and over-extracting UK business owners.
But what if, instead of economic growth, your progress myth was reducing inequality? Or growing equality within the economy?
Source: World Inequality Database wid.world
If you focused on inequality rather than economic growth, then you would find it correlates with everything we say we don’t want. Unlike economic growth, having equality as an aim actually has the advantage of having an evidence base for the claim that it improves society:
If you focused on inequality, then you would be pleased that we have had an increase in our minimum wage. You would think that the same FT article’s admission that UK graduates’ skills levels are higher than those in the United States was more important than something called a graduate earnings premium.
Burn-Murdoch is right to say asking whether university is worth it is the wrong question.
However economic growth is the wrong answer.
And I thought I would probably be stopping there for this week. But then something odd happened. A “Thought Exercise” set in June 2028 “detailing the progression and fallout of the Global Intelligence Crisis” (ie science fiction), published on 23 February, may have tanked the share price of IBM later that day. The fall definitely happened, with IBM’s share price falling 13%, its biggest fall since 2000, alongside smaller falls in other tech stocks.
Investors have recently seized on social media rumours and incremental developments by small AI companies to justify further selling, with a widely circulated blog post by Citrini Research over the weekend describing how AI could hypothetically push the US unemployment rate above 10 per cent by 2028, proving the latest catalyst.
The likelihood of the scenario portrayed is difficult to assess, but the speed with which the total economic collapse happens subsequently as described feels unlikely if not impossible. However the fact that the markets are this jittery tells us something I think. As Carlo Iacono puts it:
We are living through a period in which the gap between “plausible narrative” and “tradeable signal” has collapsed to nearly nothing. When a scenario feels real enough to model, and the underlying anxiety is already there waiting to be organised, fiction and forecast become functionally indistinguishable.
The data underlying the markets hasn’t changed, but the story has. I rest my case.
I have spent many days in rooms with groups of men (always men) anxious about their future income, where I advised them on how much to ask their companies for. Most of my clients as a scheme actuary were trustees of pension schemes of companies which had seen better days, and who were struggling to make the necessary payments to secure the benefits already promised, let alone those to come. One by one, those schemes stopped offering those future benefits and just concentrated on meeting the bill for benefits already promised. If an opportunity came to buy those benefits out with an insurance company (which normally cost quite a bit more than the kind of “technical provisions” target the Pensions Regulator would accept), I lobbied hard to get it to happen. In many cases we were too late though, the company went bust and we moved it into the Pension Protection Fund instead. That was the life of a pensions actuary in the West Midlands in the noughties. I was often “Mr Good News” in those meetings, the ironic reference to the man constantly moving the goalposts for how much money the scheme needed to meet those benefits bills. I saw my role as pushing the companies to buy out funding if at all possible. None of the schemes I advised had a company behind them which could sustain ongoing pension costs long term. I would listen to the wishful thinking and the corporate optimism, smile and push for the “realistic” option of working towards buy out.
Then I went to work at a university, and found myself, for the first time since 2003, a member of an open defined benefit pension scheme. It was (and still is) a generous scheme, but was constantly complained about by the university lecturers who comprised most of its membership. I didn’t see any way that it was affordable for employers which seemed to struggle to employ enough lecturers, were very reluctant to award anything other than fixed term contracts, and had an almost feudal relationship with their PhD students and post docs. Staff went on strike about plans to close the scheme to future accrual and replace it with the most generous money purchase scheme I had ever seen. I demurred and wrote an article called Why I Won’t Strike. I watched in wonder when even actuarial lecturers at other universities enthusiastically supported the strike. However, over 10 years later, that scheme – the UK’s biggest – is still open. And I gained personally from continued active membership until 2024.
Now don’t get me wrong, I still think the UK university sector is wrong to maintain, unique amongst its peers, a defined benefit scheme. The funding requirement for it has been inflated by continued accrual over the last 8 years and therefore so has the risk it will spike at just the time when it is least affordable, a time which may soon be approaching with 45% of universities already reporting deficits. However the strike demonstrated how important the pension scheme was to staff, something the constant grumbling before the strike had led university managers to doubt. And, once the decision had been made to keep the scheme open to future accrual, I had no more to add as an actuary. Other actuaries had the responsibility for advising on funding, in fact quite a lot of others as the UCU was getting its own actuarial advice alongside that the USS was getting, but my involvement was now just that of a member, just one with a heightened awareness of the risks the employers were taking.
The reason I bring this up is because I detected something of the same position as my lonely one from the noughties amongst the group of actuaries involved in the latest joint report from the Institute and Faculty of Actuaries and the University of Exeter about the fight to maintain planetary climate solvency.
It very neatly sets out the problem, that the whole system of climate modelling and policy recommendations to date has been almost certainly underestimating how much warming is likely to result from a given increase in the level of carbon dioxide in the atmosphere. Therefore all the “carbon budgets” (amount we can emit before we hit particular temperature levels) have been assumed to be higher than they actually are and estimates for when we exhaust them have given us longer than we actually have. This is due to the masking effects of particulate pollution in the air, which has resulted in around 0.5C less warming than we would otherwise have had by now. However, efforts to remove sulphur from oil and coal fuels (themselves important for human health) have acted to reduce this aerosol cooling effect. The goalposts have moved.
An additional reference I would add to the excellent references in the report is Hansen’s Seeing the Forest for the Trees, which concisely summarises all the evidence to suggest the generally accepted range for climate sensitivity is too low.
So far, so “Mr Good News”. And for those who say this is not something actuaries should be doing because they are not climate experts, this is exactly what actuaries have always done. We started the profession by advising on the intersection between money and mortality, despite not being experts in any of the conditions which affected either the buying power of money or the conditions which affected people’s mortality. We could however use statistics to indicate how things were likely to go in general, and early instances of governments wasting quite a lot of money without a steer from people who understood statistics got us that gig, and a succession of other related gigs over the years ahead.
The difficult bit is always deciding what course of action you want to encourage once you have done the analysis. This was much easier in pensions, as there was a regulatory framework to work to. It is much harder when, as in this case, it involves proposing changes in behaviour which are ingrained into our societies. If university lecturers can oppose something that is clearly not in the long term financial interests of their employers and push for something which makes their individual employers less secure, then how much more will the general public resist change when they can see no good reason for it.
And in this regard this feels like a report mostly focused on the finance industry. The analogies it makes with the 2008 financial crash, constant comparisons with the solvency regulatory regimes of insurers in particular and even the framing of the need to mitigate climate change in order to support economic growth are all couched in terms familiar to people working in the finance sector. This has, perhaps predictably, meant that the press coverage to date has mostly been concentrated in the pension, insurance and investment areas:
However in the case of the 2008 crash, the causes were able to be addressed by restricting practices amongst the financial institutions which had just been bailed out and were therefore in no position to argue. Many of those restrictions have been loosened since, and I think many amongst the general public would question whether the decision to bail out the banks and impose austerity on everyone else is really a model to follow for other crises.
The next stage will therefore need to involve breaking out of the finance sector to communicate the message more widely, perhaps focusing on the first point in the proposed Recovery Plan: developing a different mindset. As the report says:
This challenge demands a shift in perspective, recognising that humanity is not separate from nature but embedded in it, reliant on it and, furthermore, now required to actively steward the Earth system. To maintain Planetary Solvency, we need to put in place mechanisms to ensure our social, economic, and political systems respect the planet’s biophysical limits, thus preserving or restoring sufficient natural capital for future generations to continue receiving ecosystem services…
…The prevailing economic system is a risk driver and requires reform, as economic dependency on nature is unrecognised in dominant economic theory which incorrectly assumes that natural capital is substitutable by manufactured capital. A particular barrier to climate action has been lobbying from incumbents and misinformation which has contributed to slower than required policy implementation.
By which I assume they mean this type of lobbying:
And this is where it gets very difficult, because actuaries really do not have anything to add at this point. We are just citizens with no particular expertise about how to proceed, just a heightened awareness of the dangers we are facing if we don’t act.
But we can also, as the report does, point out that we still have agency:
Although this is daunting, it means we have agency – we can choose to manage human activity to minimise the risk of societal disruption from the loss of critical support services from nature.
This point chimes with something else I have been reading recently (and which I will be writing more about in the coming weeks): Samuel Miller McDonald’s Progress. As he says “never before have so many lives, human and otherwise, depended on the decisions of human beings in this moment of history”. You may argue the toss on that with me, which is fine, but, in view of the other things you may be scrolling through either side of reading this, how about this for a paragraph putting the whole question of when to change how we do things in context:
We are caught in a difficult trap. If everything that is familiar is torn down and all the structures that govern our day-to-day disintegrated, we risk terrible disorder. We court famines and wars. We invite power vacuums to be filled by even more brutal psychopaths than those who haunt the halls of power now. But if we don’t, if we continue on the current path and simply follow inertia, there is a good chance that the outcome will be far worse than the disruption of upending everything today. Maintaining status-quo trajectories in carbon emissions, habitat destruction and pollution, there is a high likelihood of collapse in the existing structure anyway. It will just occur under far worse ecological conditions than if it were to happen sooner, in a more controlled way. At least, that is what all the best science suggests. To believe otherwise requires rejecting science and knowledge itself, which some find to be a worthwhile trade-off. But reality can only be denied for so long. Dream at night we may, the day will ensnare us anyway.
One thing I never did in one of those rooms full of anxious men was to stand up and loudly denounce the pensions system we were all working within. Actuaries do not behave like that generally. However we have a senior group of actuaries, with the endorsement of their profession, publishing a report that says things like this (bold emphasis added by me):
Planetary Solvency is threatened and a recovery plan is needed: a fundamental, policy-led change of direction, informed by realistic risk assessments that recognise our current market-led approach is failing, accompanied by an action plan that considers broad, radical and effective options.
This is not a normal situation. We should act accordingly.
