As Governor of Britannia, Magnus Maximus has a huge army at his disposal, which is just what he needs to secure the Roman imperial throne. But perhaps the impressive general should have looked into the past before focusing on his future. Tim is joined by Paul Cooper, host of Fall of Civilizations Podcast, to explore why powerful civilisations such as the Assyrians, the Han dynasty and the Roman Empire all ultimately collapsed.

Paul Cooper is the author of Fall of Civilizations: Stories of Greatness and Decline.

For ad-free episodes, monthly bonus episodes and other benefits, check out the Cautionary Club.

[Apple] [Spotify] [Stitcher]

Me: “How do you plan to get to the start line?”

He: “I’ve asked ChatGPT. It says Elizabeth Line to Liverpool Street, then the train to Blackheath.”

That didn’t sound right. Was there a train from Liverpool Street to Blackheath? Google Maps and Citymapper suggested getting to Blackheath from Charing Cross or Waterloo.

Me: “Are you sure? I’d suggest the Circle or Bakerloo to Charing Cross.”

He frowned for a moment and pulled out his phone. “No, ChatGPT says that ‘The Circle Line is not a good choice on marathon day. It will be too crowded. There are too many stops and too many steps. It’s a route for tourists, not for runners.’”

I checked Google Maps. Sure enough, there is no train from Liverpool Street to Blackheath. ChatGPT’s recommendation would leave him stranded, trying to catch a bus over the marathon route, then trying to get on to the train from Charing Cross at a busy London Bridge. I told him that sounded like a bad idea. He frowned again and typed another query into his phone. “Oh, you’re right. ChatGPT says, ‘Correction: take the Elizabeth Line straight to London Bridge.’”

Me: “The Elizabeth Line doesn’t go to London Bridge.”

You’ve heard tales of artificial intelligence hallucinations before, but it’s not the AI that fascinates here: it’s the human. 

The route-finding algorithm on Google Maps is a minor miracle. It will solve a complex optimisation problem across multiple modes of transport, taking into account real-time congestion or delays, and it’s been available on smartphones and browsers for years. It is a proven, practical example of AI in action. So on marathon day, when the stakes are high and the clock is ticking, why would anyone turn instead to a fancy word-guessing machine such as ChatGPT?

Perhaps it’s that ChatGPT seems so human. It served up an uncanny impersonation of a friendly and knowledgeable local guide. The Circle Line? Pfft, it’s fine for tourists but you’re a marathon runner: think about all those steps! (It’s true, the creaky old Circle Line does have steps.) 

Part of the bot patter reminded me of clickbait ads: INSURANCE COMPANIES HATE THIS LOOPHOLE! ChatGPT wasn’t just giving a route, but giving a rationale, even explaining why we shouldn’t listen to the lamestream advice of Google Maps. This is the approach of a confidence trickster.

In the introduction to her book The Confidence Game, psychologist Maria Konnikova explains: “The true con artist doesn’t force us to do anything: he makes us complicit in our own undoing . . . we believe because we want to.” One difference between the con artist and the large language model (LLM) is that the con artist knows the truth and is trying to conceal it. One similarity between the con artist and the LLM is that both of them have perfected seeming plausible.

A recent paper in Nature finds that when LLMs are trained to be warm and friendly, they also produce dramatically less accurate answers, “promoting conspiracy theories, providing inaccurate factual information and offering incorrect medical advice”.

That sounds bad. I’d suggest that the reality is worse: the sycophantic AI not only produces mistakes, it persuades us to believe them. 

In 1950 Alan Turing, the mathematician and visionary of the computer age, famously proposed an “imitation game” in which a human judge would communicate through a teleprompter with a human and a computer. The computer’s job was to imitate human conversation convincingly enough to persuade the judge. 

Turing’s test remains intriguing, but there is a longstanding difficulty: the fallibility of the judge. A primitive 1960s chatbot, Eliza, responded like a parody of a therapist (“How does that make you feel?” “Why do you feel sad?” “Please go on.”). People lapped it up; it’s nice to feel listened to. A 1980s chatbot, MGonz, just fired off insults and was perfectly plausible, partly because insults are simple to deliver and mostly because they prompt rage rather than reflection in the human recipient. And Robert Epstein, an expert in the Turing Test, has written entertainingly about how he was fooled into a four-month correspondence with a sexy Russian lady who was, in fact, a 2006-era chatbot. None of these bots had a thousandth of the sophistication of a modern LLM, but they didn’t need it: when humans are sad, angry or amorous, we aren’t very sophisticated judges, either.

We are all going to find ourselves in strange variations of the Turing Test in years to come, and I wonder if we are up to it. And not just us, but those with power over us. As Cory Doctorow, author of Enshittification, is fond of observing: you won’t be replaced because an AI can do your job, you’ll be replaced because an AI salesman convinces your boss that it can. If my journey to the marathon start line is any guide, that salesman will have an easy job.

The capabilities of modern AI are impressive. But what determines whether we use it is not the capability, but the impressiveness. They are correlated but they are not the same thing. There’s a tale about the French poet Jacques Prévert seeing a fellow begging for change on the streets of Venice with a sign that read “Blind man without a pension”.

Prévert stopped to chat to him; not many people were moved to contribute, and Prévert offered to write a new sign.

The next day, he returned to find the man overjoyed. “It’s incredible; I’ve never received so much money in my life.” 

Prévert had written: “Spring is coming, but I won’t see it.” 

The new sign contained no news — in fact, it was less informative than the old. But it told a story. Google Maps was the first sign: it told me where to get my train. ChatGPT was the second sign: it told my companion not just where to go, but how to feel about taking such a clever route.

I left him at Paddington, urging him not to try to take the non-existent Elizabeth Line train to London Bridge. I am not sure I was as convincing as ChatGPT.

I ran the London Marathon in support of the Teenage Cancer Trust – not too late to make a donation.  

Written for and first published in the Financial Times on 6 May 2026.

It’s D-Day and the Allies are about to invade Nazi-occupied France. For the landings to succeed, American soldiers on Omaha Beach will have to break through some formidable coastal defences – Hitler’s Atlantic Wall. Sherman tanks will come in very handy – and the Allies have come up with a novel solution for getting them to the beach. These tanks will swim. Everyone from Winston Churchill down thought swimming tanks were a great idea… but were they?

This episode was previously available to members of the Cautionary Club, and Pushkin+ subscribers.

[Apple] [Spotify] [Stitcher]

Further Reading

The Second World War. Vol II. W Churchill.

Normandy 44 by James Holland

Overlord by Max Hastings

Beyond the Beachhead by Joseph Balkoski

The Maritime Archaeology of Duplex Drive Tanks in the United Kingdom by Thomas Cousins, Thomas Harrison and Dave Parham. Bournemouth University, UK

The Design of Everyday Things by Don Norman 

Clay Christensen’s Milkshake Marketing by Carmen Nobel (Harvard Business School)

Don’t Solve the Wrong Problem by Prashant Verma 

What is Workplace Innovation? By The European Workplace Innovation Network 

Interview with D-Day veterans – D-Day I Was There. BBC News.

This is not wholly to our collective advantage. Emotions often lead us astray. The behavioural economist George Loewenstein, in a 1996 paper titled “Out of Control: Visceral Influences on Behavior”, noted that extreme visceral emotions often lead to self-destruction. There is the alcoholic who is willing to abandon her children for the sake of the next drink; the suspect in a long interrogation who will sacrifice years of freedom for a glass of water and the chance to sleep; the arachnophobe who will risk injury in a scramble to escape from a toy spider that he knows perfectly well cannot hurt him; the road-raging driver who knows, somehow, that he is doing himself no favours even as he swings the punch. In the grip of emotion, we are not our best selves.

We are also often unreflective about this fact. “People underweigh, or even ignore, visceral factors that they will experience in the future, have experienced in the past, or that are experienced by other people,” writes Loewenstein.

Hard experience suggests this is true, but we also have some light-hearted experimental evidence on the visceral emotion of gluttony. Daniel Read and Barbara van Leeuwen ran a study in which participants chose between snacks ranging from apples to Mars bars. If the snack was for immediate consumption, people chose the chocolate bar. If told they were pre-ordering for next week’s follow-up, they chose fruit. A week later, offered a chance to change their minds, they caved and grabbed the Mars bar again.

The point — familiar but deep — is that we are consistently inconsistent, intending good things for the future, surrendering to our animal selves when the future arrives, and then somehow expecting to do better next time. What is true for snacks may also be true for how we choose to engage in politics. In principle, favouring the apple of thoughtful discussion, while in practice seizing the Mars bar of furiously retweeting angry influencers.

Intuitively, it seems that politics is more emotionally fraught than it was a generation ago. Is that right — and if so, why? A recent working paper from Eva Davoine, Stefanie Stantcheva, Thomas Renault and Yann Algan trawled through posts on X. They used a large language model to evaluate the emotional content of tweets from US citizens on political subjects including abortion, immigration, tax and inequality, and democracy itself. (Stantcheva and her colleagues went to some lengths to check that the people behind the accounts really exist.)

Two findings immediately emerged. The first is that of all the emotions expressed in tweets, by far the most common is anger. Other emotions, positive and negative (fear, disgust, gratitude, hope, joy) barely appear, but anger is everywhere.

The second is that anger is on the rise. The data starts in 2013, when Barack Obama was president, and for years about 25 per cent of political tweets were angry, and about two-thirds expressed no clear emotion. Anger started rising in 2016 and by 2020, between 40 and 50 per cent of political tweets were angry. That is partly because some calm tweeters have quit and been replaced by some angry ones, but mostly it is the same people as ever, tweeting more angrily than they once did.

Who are the angriest tweeters? People at the political extremes, both left and right; people who follow a lot of politicians and political influencers; and, interestingly, people over the age of 65. The boomers are furious, at least the ones doing politics on social media. (Women express more anger than men, and Republicans more than Democrats, but, in both cases, the margins are small.)

The researchers also look at the “supply” of anger from politicians and find, again, large increases since 2013. They also find strong political cycles (politicians in opposition are much more likely to express anger) and that while there was a sharp rise in angry tweets from politicians, there was a far more muted rise in angry speech on the floor of Congress.

That contrast between social media and traditional speechmaking is instructive. It’s quite possible that people are angry in response to real problems, and it’s also possible that people are angry because the defining political figure of the age, Donald Trump, both uses and generates anger.

But it may also be that anger is on the rise because the medium of political discussion encourages it. Stantcheva and her colleagues find that angry tweets by citizens are nearly 90 per cent more likely to be retweeted than non-emotional tweets, while angry tweets from congressional politicians are nearly 50 per cent more likely to be retweeted. If social media rewards anger, people will respond to the incentive.

This may also explain the rise of populism. A recent study by George Ward, Sandra Matz and others found a correlation between negative emotions — fear, anger, depression, sadness — and support for populists and populist causes including Brexit in 2016 and Donald Trump in 2016 and 2020. The causal chain is too stretched to claim confidently that social media caused Brexit and Trump, but it is perfectly plausible to suggest that Twitter was fertile soil for populist seeds to sprout.

There is nothing new about anger in politics. “Before the Hate had proceeded for thirty seconds,” wrote George Orwell in Nineteen Eighty-Four (1949), “uncontrollable exclamations of rage were breaking out from half the people in the room. The self-satisfied sheep-like face on the screen, and the terrifying power of the Eurasian army behind it, were too much to be borne.”

Orwell wasn’t just trying to imagine the future, but refracting the vicious propaganda of the 1930s and 1940s. He knew very well that we are not our best, most reflective selves when we are angry — ruthless politicians will exploit that.

George Loewenstein argued not just that we are often powerless in the face of strong emotions, but that we are in denial: we tend to tell ourselves that visceral feelings will not get the better of us, when all too often they do.

Better to avoid temptation. A well-stocked bar is no place for a recovering alcoholic. It’s always easier to diet if the Mars bar is in a far away supermarket rather than the kitchen shelf. And if you’re interested in politics, John Stuart Mill’s On Liberty is always worth a read, but at all costs stay away from Bluesky and X.

Written for and first published in the Financial Times on 29 April 2026.

Loyal readers might enjoy How To Make The World Add Up.

“Nobody makes the statistics of everyday life more fascinating and enjoyable than Tim Harford.”- Bill Bryson

“This entertaining, engrossing book about the power of numbers, logic and genuine curiosity”- Maria Konnikova

I’ve set up a storefront on Bookshop in the United States and the United Kingdom. Links to Bookshop and Amazon may generate referral fees.

Thomas Midgley’s inventions caused his own death, hastened the deaths of millions of people around the world, and very nearly extinguished all life on land. 

Midgley and his employers didn’t set out to poison the air with leaded gasoline or wreck the ozone layer with CFCs – but while these dire consequences were unintended… could they have been anticipated? 

For ad-free listening and bonus episodes, video conversations and our newsletter, join the Cautionary Club.

[Apple] [Spotify] [Stitcher]

Further reading and listening

On Thomas Midgley

Charles Kettering “Biographical Memoir of Thomas Midgley Jr”

Fred Pearce “Inventor hero was a one-man environmental disaster” New Scientist 7 June 2017

Edelmann, F.T. 2016 (31:viii): The life and legacy of Thomas Midgley Jr. Papers and Proceedings of the Royal Society of Tasmania 150(1): 45–49.

On CFCs and the Ozone Layer

Sharon Roan Ozone Crisis

Press Release on the award of the 1995 Nobel Prize in Chemistry to Paul Crutzen, Mario Molina, and F Sherwood Rowland

New York Times obituary of Joseph Farman

On Lead

Mike Sutton “Pb or not Pb: the toxic question of leaded fuel” Chemistry World 20 December 2021

Deceit and Denial: The Deadly Politics of Industrial Pollution, Gerald Markowitz and David Rosner, 1987

UNEP press release on leaded petrol

Tom Whipple “Leaded Petrol reduces intelligence decades later” The Times March 2017

On Unanticipated Consequences

Robert K Merton The Unanticipated Consequences of Purposive Social Action. American Sociological Review, Vol. 1, No. 6 (Dec., 1936), pp. 894-904

Frank de Zwart, Unintended but not Unanticipated Consequences, Theory and Society 44(3), April 2015

Nitin Nohria and Hemant Taneka “Managing the Unintended Consequences of Your Innovations” Harvard Business Review 19 January 2021

Two pathologists, Eberhard Schairer and Erich Schöniger, published a study of “lung cancer and tobacco consumption” well before Doll and Hill took up the question. Schairer and Schöniger began by noting that lung cancer, a rare disease in the 19th century, had shown a “pronounced increase”, and then dismissed the plausible-seeming idea that the cause was vehicle exhausts. Lung cancer was on the rise in rural areas just as much as urban, they noted, adding that “the male gender is much more frequently afflicted . . . than the female,” yet “both genders are exposed to almost the same degree” to exhaust fumes. More plausible, they suggested, was that cigarettes were to blame.

Schairer and Schöniger sent out questionnaires to the families of people who had died from cancer, and also to living men in their fifties (at peak risk of lung cancer), enquiring about their smoking habits. They found a strong correlation between being a heavy smoker and developing lung cancer, but no such link between smoking and stomach cancer. Schairer and Schöniger never pretended to have the last word on the subject, they thought their findings were not certain but “only likely”. It was a small study and there are question marks over the survey methods used. But decades later, Richard Doll himself described the study as “percipient” and “important” even if it fell short of decisive proof.

Why, then, is the study not better known? The overwhelming weakness in Schairer and Schöniger’s study is simple: they were German, writing in German, and they published in a German scientific journal in 1943. The immediate consequence is that Doll and Hill didn’t see it. Copies of the study didn’t reach the UK at all during the war.

But the problem for Schairer and Schöniger was not a language barrier or a failure of scientific communication because of the strictures of war. It was that their entire enterprise was fatally tainted by association with the Third Reich. Doll notes that in postwar West Germany, the study was ignored and that “interest in the effects of smoking waned, as a reaction to the anti-smoking policies of the Nazi government”.

Those policies are striking indeed. Hitler was a vegetarian — a fact so well known as to have become a punchline — but he was also a non-smoker. Nazi anti-tobacco activists were fond of pointing out that Mussolini and Franco also abstained, while Churchill, Roosevelt and Stalin all smoked heavily. Nazi cartoons depicted grotesque caricatures of Jews and dark-skinned people drinking and smoking, connecting two Nazi obsessions: their fear of racial impurity and their fear of impure substances such as alcohol, tobacco and processed food. Some of the Nazi ideas about health, clean living and being ever vigilant about toxins feel unnervingly fashionable.

The Third Reich pursued its promotion of healthy lifestyles with some good old-fashioned public health messaging with an overtone of “do it for the good of the nation” that is probably much more sinister with the benefit of hindsight. Inspiring posters pointed out that healthy Aryan women drank apple juice and did not smoke so they could produce healthy babies, while hard-working men could save so much money from quitting smoking that they could buy a Volkswagen.

For chief propagandist Goebbels, there was a risk of taking this campaign against smoking too far. Many teachers smoked, and doctors, and for that matter many senior Nazis, including Hermann Göring. The propaganda ministry feared that a hardline wartime campaign against smoking would inevitably fail and the government would look foolish.

Yet Hitler clearly felt quite passionate about the issue. In 1941, Karl Astel, a professor of medicine as well as a prominent racial hygienist, established the Scientific Institute for Research into the Hazards of Tobacco at Friedrich-Schiller University in Jena. Hitler sent a telegram of congratulations and 100,000 Reichsmarks from his personal office, a substantial sum.

Astel was a strong supporter of Hitler as early as 1923, and received the “Golden Badge of Honour” for being one of the early members of the Nazi party. He was intimately involved in organising the Holocaust and even more closely associated with the murder of 200,000 disabled people. Shortly before the end of the war, and presumably contemplating being tried for war crimes, he killed himself.

Astel also funded Schairer and Schöniger’s research.

So how complicit were Schairer and Schöniger themselves in the evils of the Third Reich? We know little about Schöniger, but we do know that Schairer joined several Nazi organisations, including — along with 10 million others — the Nazi party. In an obituary essay, Schairer’s son declared that his father “never was a member of the resistance and he was not a hero” but that he was a “liberal person” who regarded the Nazi manifesto as “inhuman”.

While it is only natural to want to know what Schairer and Schöniger thought and did about Hitler, that has very little to do with what they thought and did about the link between cigarettes and lung cancer. Their research was important and should have been influential, but it was ignored. At a distance of more than eight decades it’s easy to shrug; we figured out the link between cigarettes and cancer a few years later, after all.

But the delay was real and deadly, especially in West Germany itself. Historian Robert Proctor, author of The Nazi War on Cancer, speculates that in postwar Germany the Nazi campaign against tobacco delayed “the development of effective tobacco measures by several decades”. Public health campaigners and epidemiologists have always had powerful enemies in the form of Big Tobacco. But who needs an enemy if your loudest cheerleader is Adolf Hitler?

For all our lofty aspirations to scientific rigour, we are social creatures, heavily influenced by the beliefs of those we admire and those we despise. But even those we despise are not deluded about everything. The Nazis were so monstrously wrong about so many things that it is hard to imagine that they were ever right. Life is rarely that simple.

Written for and first published in the Financial Times on 22 April 2026.

Loyal readers might enjoy the book that started it all, The Undercover Economist.

I’ve set up a storefront on Bookshop in the United States and the United Kingdom. Links to Bookshop and Amazon may generate referral fees.

Part Two: When Tudor polymath John Dee meets a man who claims he can speak with angels, his path to understanding the universe suddenly becomes clear. At their instruction, the pair begin searching for the fabled philosopher’s stone. But the angels grow increasingly demanding, and soon Dee must confront a terrible ultimatum.

Centuries later, a strange incident in a French town suggests that angels may still be with us.

For ad-free listening and bonus episodes, video conversations and our newsletter, please consider joining the Cautionary Club.

[Apple] [Spotify] [Stitcher]

A key source was The Diaries of John Dee (1998), edited by Edward Fenton. Several other books were very useful:

John Dee’s Conversations with Angels: Cabala, Alchemy and the End of Nature (1999) by Deborah Harkness

The Years of the Wizard: The Strange History & Home Life of Renaissance Magicians (2025) by Rachel Morris

The Arch-Conjuror of England (2011) by Glyn Parry

The Secrets of Alchemy (2012) by Lawrence M. Principe

Prophecy: Prediction, Power and the Fight for the Future, from Ancient Oracles to AI (2026) by Carissa Véliz

The Queen’s Conjuror: The Life and Magic of Dr Dee (2002) by Benjamin Woolley

The following websites and articles also helped us tell John Dee’s story:

“Phantom pregnancy: The mental health condition that mimics a baby’s arrival” by Rosemary Counter, National Geographic (11 September 2023)

“Some Workmen’s Wages in 1588” by Elizabethan.org

“A robot in Prague and an elixir for Rudolph II” by Habsburger.net 

“John Dee: Elizabethan 007, scientist, magician and spy” by History Extra (8 October 2021)

“Nothing But Solitude” by Christopher P. Heuer, Lapham’s Quarterly (14 May 2019)

“Mathematics, navigation and empire” by Alex Grover, Royal Museums Greenwich (8 July 2019)

“Martin Frobisher’s North West Passage expedition 1576–78” by Royal Museums Greenwich

“The Coronation of Queen Elizabeth” by A. L. Rowse, History Today (May 1953)

“Elizabeth I’s Coronation Procession” by Tudor Times (16 August 2019)

“John Dee and Edward Kelly: Through a Glass Darkly” by Michael Wilding, The Brazen Head (18 October 2020)

I previously wrote about predictions here and about Dan Kahan’s work here. For more on “badges of membership” see “Climate-Science Communication and the Measurement Problem”, Advances in Political Psychology (20 February 2015) by Dan M. Kahan and “Identity: Dan Kahan” by Informal Science.

For the story of Gilles D’Ettore, we drew on the podcast series The Mystic and the Mayor (2025), as well as the following articles:

“French town reels from fortune teller scandal” by Chris Bockman, BBC (27 May 2024)

“«Se faire flouer ainsi par une prétendue voyante…» : l’affaire du maire d’Agde «ensorcelé» sidère” by Christian Goutorbe, Le Parisien (23 March 2024)

“Comment le maire d’Agde est tombé sous l’emprise d’une voyante ventriloque et la voix de « l’archange Michaël »” by Samuel Laurent, Le Monde (18 April 2024).

“France’s ‘swinger’ capital rocked by fortune teller scandal” by Harriet Marsden, The Week (4 June 2024)

It was a testament to the persuasive power of an apolitical nerd. The audience realised that this insurance analyst had no interest in the Woke Dems — he was just describing the world as he saw it. But it was also a window into the way that our views about the world shape financial markets for risk, and financial markets for risk shape our views about the world.

There are two venerable insurance industries, the French economist Michel Albert once explained. One has its roots in Alpine pastures, where Swiss villagers agreed to help each other out if one farmer’s cow died. The other was born in Edward Lloyd’s coffee house in 17th-century London, where sailors and shipowners would gather to gamble on which ships would sink and which would return safely to port.

Today we call both of these activities “insurance”, but they have very different souls. The Swiss version is all about mutual assistance, figuring out who is part of a community and who is not, and sharing burdens. The London version begins with the insight that some risks are fun, but also that people like to be smart about those risks.

Edward Lloyd himself created a network of well-informed correspondents across the ports of Europe and published a newsletter specialising in maritime cargo and foreign affairs. Cuthbert Heath, a Lloyd’s underwriter from the late 19th century onwards, went on to specialise in acquiring the data needed to sell insurance against losses from earthquakes and hurricanes. Selling well-priced insurance is a lucrative business, and the better the information and judgment about future risks, the better the business will be.

Financial risk contracts have long provided incentives to improve the state of our knowledge. In the 16th century the inveterate gambler Girolamo Cardano revolutionised our understanding of probability, while in the 17th the comet guy Edmond Halley used data on births and deaths gathered in Breslau to demonstrate that the government was selling life annuities too cheaply. A decade ago, I saw an agricultural insurance expert tell agribusiness traders something they didn’t know about climate change. It is all part of the same process.

With this noble tradition in mind, perhaps we should not be surprised that prediction markets are taking off. They are not new; in 2003 the Pentagon had been pondering a “Policy Analysis Market” in which people would trade on the risk of — say — a deadly terrorist strike, or, um, attacks on shipping in the Strait of Hormuz. The idea that the US intelligence community might host such a market was a political non-starter and the plan was dropped.

Yet the controversial idea made some sense. Prediction markets collect information by offering money for it. Markets may not be perfectly efficient but they are often informative: if you want to understand whether Arsenal’s lead in the Premier League table is commanding or precarious, the betting odds are a good place to start. Prediction markets, like any betting market, aggregate information.

The mood music, and the rules on gambling, have changed since 2003. Prediction markets such as Kalshi and Polymarket have become well known for offering contracts on all sorts of events, ranging from “Will Jerome Powell be arrested in 2025?” to degenerate nonsense such as “Will Jerome Powell say ‘Good afternoon’ during his December 2025 press conference?”

The libertarian in me says that people should be allowed to bet on pretty much anything, but the economist in me has some concerns. Of course, there is a worry — or there should be — that some people find gambling ruinously addictive, especially with smartphones putting a slot machine in every pocket.

But we should also be concerned that in trying to predict the future, we change it for the worse. It might indeed be useful to get a sense of whether Jerome Powell will be arrested, because it matters if he is. It does not matter whether he says “Good afternoon”, so there is little benefit in such a market existing. The market in “Good afternoon” does, however, create the clear risk that somebody tries to bribe or threaten Powell.

That is probably the least of his worries, but it is not hard to find examples of betting markets corrupting the real world. The most obvious cases are of crooked sports bets, where athletes rig games for their own benefit — or the benefit of people who are bribing or threatening them. It’s not even necessary to throw a match: it’s possible to bet on all sorts of trivia that are only peripherally related to the result.

Or consider Emanuel Fabian, The Times of Israel journalist who was offered bribes, then death threats, to change his report that an Iranian missile had struck near Jerusalem on March 30. A great deal of money was — courtesy of Polymarket — riding on the question of whether Iran had or had not succeeded in striking Israel. Without his consent, Fabian found himself forced into the perilous role of refereeing which side had won the bet.

Fabian’s plight is one of those things that, with hindsight, was obviously bound to happen. Whether prediction markets mean that athletes are being threatened if they don’t change a game, or journalists are being threatened if they don’t change a news report, this all seems like something that could be better thought through.

It would not be the first time that a new, disruptive industry smacked headlong into a problem that the stodgy old incumbents have understood for a very long time. In the early 1960s, life insurers were alarmed to discover that they were paying vast sums in compensation for “loss of limb” to policyholders in the Florida panhandle. One fellow lost a foot; fortunately he had a tourniquet with him . . . and insurance from several dozen different companies. Another chap bought insurance then shot off his own foot just 12 hours later “while aiming at a squirrel”.

There’s an old-fashioned phrase for this: moral hazard. And perhaps some old-fashioned caution is in order. Betting on the future can protect us from risk, and it can make us smarter about risk too. But it can also corrode and corrupt. Let’s be careful.

Written for and first published in the Financial Times on 15 April 2026.

In Tudor England, the line between mathematics and the mystic arts is vanishingly thin. Straddling both worlds is John Dee, a brilliant scholar and astrologer whose intellect grants him access to the highest circles of power. Dee navigates the politics of the court by making bold prophecies, which win him royal favour. But even correct predictions may come with a price – and laying claim to the future is a dangerous game. 

For ad-free listening and bonus episodes, video conversations and our newsletter, please consider joining the Cautionary Club.

[Apple] [Spotify] [Stitcher]

A key source was The Diaries of John Dee (1998), edited by Edward Fenton. Several other books were very useful:

John Dee’s Conversations with Angels: Cabala, Alchemy and the End of Nature (1999) by Deborah Harkness

The Years of the Wizard: The Strange History & Home Life of Renaissance Magicians (2025) by Rachel Morris

The Arch-Conjuror of England (2011) by Glyn Parry

The Secrets of Alchemy (2012) by Lawrence M. Principe

Prophecy: Prediction, Power and the Fight for the Future, from Ancient Oracles to AI (2026) by Carissa Véliz

The Queen’s Conjuror: The Life and Magic of Dr Dee (2002) by Benjamin Woolley

The following websites and articles also helped us tell John Dee’s story:

“Phantom pregnancy: The mental health condition that mimics a baby’s arrival” by Rosemary Counter, National Geographic (11 September 2023)

“Some Workmen’s Wages in 1588” by Elizabethan.org

“A robot in Prague and an elixir for Rudolph II” by Habsburger.net 

“John Dee: Elizabethan 007, scientist, magician and spy” by History Extra (8 October 2021)

“Nothing But Solitude” by Christopher P. Heuer, Lapham’s Quarterly (14 May 2019)

“Mathematics, navigation and empire” by Alex Grover, Royal Museums Greenwich (8 July 2019)

“Martin Frobisher’s North West Passage expedition 1576–78” by Royal Museums Greenwich

“The Coronation of Queen Elizabeth” by A. L. Rowse, History Today (May 1953)

“Elizabeth I’s Coronation Procession” by Tudor Times (16 August 2019)

“John Dee and Edward Kelly: Through a Glass Darkly” by Michael Wilding, The Brazen Head (18 October 2020)