Michael Wooldridge is like the teacher you wish you’d had: easy to talk to, great at breaking down tough ideas into simple terms, not too intellectual or trying too hard to be cool, and genuinely excited about his work. “I love it when you see the light go on in someone, when they grasp something they didn’t understand before,” he says. “I find that incredibly rewarding.”
He comes across as a regular guy, which, as an Oxford professor with over 500 scientific papers and 10 books to his name, he clearly isn’t. Typically, his favorite work is his contribution to Ladybird’s Expert Books—an update of the classic children’s series—on artificial intelligence. “I’m very proud of this,” he says, handing me a copy from his bookshelf. We’re in his study at the University of Oxford’s somewhat ordinary computing department on a sunny spring day. Maybe it’s the campus setting, but our conversation feels almost like a seminar.
Wooldridge is a skilled public speaker, especially on artificial intelligence—a field he’s worked in for over 30 years, but one he still approaches with a healthy dose of skepticism. In his 2023 Christmas lectures for the Royal Institution, titled The Truth about AI, he brought in a robotic dog and asked his school-age audience to vote on whether they’d hit it with a baseball bat. And to explain reinforcement learning, he recreated the classic 80s movie WarGames, where a young Matthew Broderick prevents a nuclear disaster by getting the US military computer to play tic-tac-toe with itself (until it realizes there’s no real way to win). “Matthew Broderick was in London at the time. We tried to get him to come to the Christmas lecture, but he couldn’t make it,” says Wooldridge. “So we named our computer BrodeRick in his honor.”
WarGames is actually pretty close to the topic of Wooldridge’s latest book, Life Lessons from Game Theory: The Art of Thinking Strategically in a Complex World. He’s been teaching this subject to his students for over 15 years, he says. Now it’s our turn. There’s no math in Wooldridge’s book; instead, he turns game theory into 21 easy-to-understand scenarios, covering everything from Atlantic cod fishing to Pepsi vs. Coca-Cola to the existence of God.
“It’s surprising how many global events can be explained by a relatively small number of game theory models,” Wooldridge says. One of the simplest is the game of “chicken,” which he illustrates in his book using a scene from the James Dean movie Rebel Without a Cause (none of his students had heard of it, he admits). Two teenagers drive their cars toward a cliff; the first to jump out is the “chicken” and loses. If they both jump at the same time, it’s a draw; if neither jumps, you lose badly (spoiler alert: that’s what happens in the movie).
The theory lesson here is about Nash equilibriums (we won’t get into the details)—but in practice, we see this game playing out in real life all the time. The Cuban Missile Crisis used to be the go-to example, but another one is unfolding right now: the US-Iran conflict. “You’ve got two sides making ever-escalating threats against each other; someone has to back down at some point,” says Wooldridge. “The danger is, if neither backs down, you pass a point of no return and get the worst-case scenario for everyone.”
Is there any way out of this? “Well, one way the game can change is if a third party steps in and offers an incentive for one side to act differently.” Another option is to bypass the game by communicating with your opponent. That’s what happened during the Cuban Missile Crisis, butIt feels less likely here. “Although, I have to say, Iran seems to be playing it much more cleverly, in the sense that the US side is very, very unpredictable. Now, being unpredictable is also a classic game theory strategy, but it makes it very hard for the other side to know how to respond. If you’re really up against an irrational player, one thing game theory says is that you just hedge your bets against the worst-case scenario.”
This isn’t just about warfare or even games, Wooldridge stresses. In his book, he defines game theory as “a mathematical theory that aims to understand situations where self-interested parties interact with each other.” He argues that this can apply to all kinds of situations: social, political, and philosophical.
View image in fullscreen
Genuinely enthusiastic … Wooldridge in 2023. Photograph: Paul Wilkinson
The idea of a “zero-sum game,” for example, has become a common term (partly thanks to WarGames), even if it’s widely misunderstood. A zero-sum game isn’t simply one where one side gains what the other loses; it’s one where the goal is to make your opponent lose as badly as possible, Wooldridge explains. So, technically, chess isn’t a zero-sum game because you’re just trying to win, not to destroy or humiliate your opponent. There’s a social and political side to this. “This zero-sum mindset is very harmful. It’s a very male trait,” he says. “And the evidence shows that not only do you not necessarily do as well in life as you could, but you actually end up more miserable. You feel like you have less control over your own affairs. One of the key lessons from game theory is that, in reality, most of the interactions we have are not zero-sum.”
This adversarial worldview drives populist politics – in the sense of “migrants are coming to take your jobs.” You’re losing because others are winning. One of Wooldridge’s favorite games encourages us to think the opposite: the Veil of Ignorance, created in 1971 by philosopher John Rawls. The idea is that you can design society however you want, but afterward, you’ll be placed randomly within it. Wooldridge calls it “a beautiful thought experiment … It encourages a socially good outcome, but people are still following their own self-interest.” He adds that Bill Clinton and Barack Obama were both fans.
It’s not immediately clear how game theory fits with AI, but these days, it’s a big part of it, Wooldridge explains, especially in his main area of interest: multi-agent systems – programs that interact with each other and act on your behalf. “So if I want to arrange a meeting with you, why would I call you up? Why doesn’t my Siri just talk directly to your Siri?” These kinds of interactions are built into our online lives. For example, online auctions like those on eBay, where you’re trying to sneak in the winning bid at the last moment. “If my agent is going to interact with your agent, and my preferences don’t necessarily match yours, then the theory that explains how you should think about those interactions is game theory.”
View image in fullscreen
A Tandy TRS-80 PC, Wooldridge’s first machine. Photograph: Photology1971/Alamy
When Wooldridge started out, AI was almost an abstract concept. He got into computing through amateur enthusiasm. Growing up in rural Herefordshire, as the son of a middle manager at the local cider company, it was a big deal when his local electronics shop had a home computer for sale, around 1980. “This sounded ridiculous because I thought computers were multimillion-pound things.” The shop owners kindly let him try it out (it was a Tandy TRS-80). “I went back week after week and taught myself to program. I was literally sitting in the shop window on the computer.” He went on to study…After studying computing as an undergraduate, I started a PhD in AI in 1989, then did an internship with Janet (the Joint Academic Network), which was basically the UK part of the early internet. Technology has moved on incredibly since then, but as Wooldridge says, “the core techniques that drove the current AI revolution were invented by the mid-80s.” He mentions Geoffrey Hinton, a pioneer of artificial neural networks – the mechanism that now powers machine learning. “The only obstacle standing in the way of the AI revolution in the 1980s, really, was that computers weren’t powerful enough and we didn’t have enough data.”
The next generation of influencers will agree to have everything they say, do, and see used for AI.
When it comes down to it, Wooldridge says, the breakthrough success of GPT-3 in 2020 was largely “based on a bet that OpenAI made that if they did the same thing, only 10 times bigger, that would deliver results. A lot of people at the time, including me, were very sceptical about it. I’m a scientist; I would like to see advances through scientific development, not just by throwing more computer power at it. But it turned out that, actually, that was a very successful bet.” Does that suggest OpenAI boss Sam Altman and his peers aren’t the tech geniuses people think they are? “I’ve never met Sam Altman; I don’t know,” he says diplomatically. “He’s clearly delivered something remarkable.”
Geniuses or not, these AI pioneers may be reaching their limits. A few years ago, people like Altman and Google DeepMind’s Demis Hassabis expected to achieve AGI – human-level artificial general intelligence – within a few years. “I personally think they’re overoptimistic,” Wooldridge says. You can talk to ChatGPT about quantum mechanics in Latin, he points out, “but at the same time, we don’t have AI that could come into your house, that it had never seen before, locate the kitchen and clear the dinner table” – something a minimum-wage human worker could do.
“The limits are the computer power and the data that you’re able to throw at it. And data is now a real constraint.” The whole of Wikipedia made up just 3% of GPT-3’s training data, he says. “Where do you get 10 times more data from next time around?” Data is becoming a valuable resource for that reason, and some organisations possess a potential treasure trove of it. “The NHS is sitting on a huge amount of data about human beings. That’s the most valuable kind of data imaginable.” Private corporations would pay a lot for it, he says, “but I suspect that whoever signed off on such a deal would live to regret it.” He imagines a dystopian future scenario where “you’re only able to have access to the NHS if you agree to be wired up to wearable tech that monitors you on a regular basis … I think we are very quickly going to a world where the next generation of online influencers basically agree to have all of their life experiences, everything they say and do and see, harvested to provide data for AI.”
From an academic standpoint, Wooldridge resents the way Silicon Valley has come to dominate the AI field, both in terms of resources (“GPT-3 required 20,000-odd AI supercomputers to train; there are probably a couple of hundred in the whole of the University of Oxford”) and the public conversation. “We have seen the narrative stolen by Silicon Valley, which is promoting a version of AI [profit-driven, job-replacing and almost entirely focused on large language models] that certainly me and an awful lot of my colleagues have no interest in promoting or building,” he says. “It’s kind of depressing, as somebody who’s spent their career trying to build AI to make a better world and to improve people’s lives.”He continues: “If you look at the big picture, AI offers a huge range of benefits that often go unnoticed because large language models dominate all the attention.” He mentions a team in Oxford developing an AI-powered tool that can analyze a heart scan from a simple ultrasound, sent to your GP via mobile phone. “This is the kind of expensive care the NHS struggles to provide, suddenly available at very low cost.”
In 2025, Wooldridge won the Royal Society’s prestigious Faraday Prize for his skill in explaining scientific ideas to the public. His lecture in February was titled This Is Not the AI We Were Promised. Around that time, he suggested AI could have a “Hindenburg moment” – the Hindenburg crash destroyed the airship industry overnight. “It’s entirely possible we could see a similar AI-related disaster,” he says. “Computer programs fail in all sorts of ways, and we’re completely dependent on a computing network where AI is increasingly embedded.” That said, when it comes to existential risks, “AI isn’t high on my list of things that keep me up at night,” he adds. “I don’t worry about a robot takeover. At least, it’s not in my top five.” The fact that he considers nuclear war a bigger threat isn’t exactly reassuring, though.
View image in fullscreen
Considering the future … Michael Wooldridge. Photograph: Philippa James/The Guardian
If he could, he would slow down AI development, “just so we have more time to understand what’s happening.” He points out it’s a classic “prisoner’s dilemma,” a key idea in game theory. In the standard scenario, two prisoners must decide separately whether to confess to a crime they committed together or stay silent. If one confesses and the other doesn’t, only the confessor goes free. If both confess, they each serve a shorter sentence. If both stay silent, they serve an even shorter sentence. So they’d be better off if both agreed to stay quiet, but neither knows what the other will do. Counterintuitively, game theory says the smartest move is to confess.
By the same logic, AI companies are locked in a race to get ahead. Their competition leads to more spending, resources, and energy-hungry data centers, with no net benefit for humanity. But here we are. “We have a small number of very wealthy companies chasing AI, while at the same time saying they’re afraid something will go horribly wrong. So why are they still chasing it? Because they think if they back down, someone else will.”
Was he ever tempted by Silicon Valley himself? “There were a few points where that could have happened, I suspect,” he says. “But I’m turning 60 this year, and it’s a young person’s game now.” Some argue there’s no point in studying anymore, since AI is predicted to replace so much human activity. Wooldridge doesn’t see it that way. “I didn’t get into computing because I thought it would give me a good job. I got into it because I was genuinely interested.” He says many parents ask him what their kids should study at university, “and the answer is: ‘Let them study something they’re really passionate about.’ I think that’s the most important thing by far.”
Life Lessons from Game Theory: The Art of Thinking Strategically in a Complex World by Michael Wooldridge is published on 21 May (Headline, £25). To support the Guardian, buy a copy at guardianbookshop.com. Do you have an opinion on the issues raised in this article? If you would like to submit a response of up to 300 words by email to be considered for publication, please do so.To have your letter published in our letters section, please click here.
Frequently Asked Questions
Here is a list of FAQs based on the topic written in a natural tone with clear direct answers
BeginnerLevel Questions
1 Wait isnt everyone terrified that AI is going to take over the world Why isnt this expert worried
Answer Professor Wooldridge says the robots taking over idea is more science fiction than reality He believes we are very far from creating a machine that has its own goals or consciousness The real danger isnt a robot rebellion its how humans use the technology
2 If we dont have to worry about a robot apocalypse what should we be worried about
Answer The biggest risks come from big tech companies controlling AI He worries about loss of privacy biased algorithms making unfair decisions and the spread of misinformation The danger isnt the AI itself but the power it gives to the people running it
3 So is AI actually good for anything or is it all bad
Answer Its not all bad Professor Wooldridge points to real benefits AI is great at specific repetitive tasks like spotting diseases in medical scans optimizing traffic flow and helping with scientific research The key is using it as a tool not a replacement for human judgment
4 What does big tech have to do with this Isnt AI just a computer program
Answer Big tech owns the massive computer power and the huge amounts of data needed to train the most powerful AI They decide how its built and who gets to use it The risk is that a few companies have too much control over a technology that affects everyone
AdvancedLevel Questions
5 The article mentions the occasional benefits Whats a specific realworld benefit that Wooldridge highlights
Answer He often points to breakthroughs in science and medicine For example AI can analyze millions of protein structures in hours a task that would take humans years This has dramatically sped up drug discovery and our understanding of diseases
6 What is the alignment problem and does Wooldridge think its a real threat
