‘I don’t worry about a robot takeover’: Oxford AI expert Michael Wooldridge on Big Tech's real threats, and surprising wins

Ask most Americans what worries them about artificial intelligence, and the answers tend to cluster around the same thing: job losses, deepfakes, or the classic Hollywood scenario of machines taking over. However, Michael Wooldridge, a professor of computer science at the University of Oxford, who has spent more than 30 years studying AI, says we’re worrying about the wrong things altogether.

“I don't worry about a robot takeover,” he says matter-of-factly. Then, what is keeping him up at night? How a handful of massively powerful companies are wasting data, money, and energy in a race that may not actually benefit any of us.

Wooldridge is the author of Life Lessons from Game Theory, a book that applies the logic of strategic decision-making to explain everything from geopolitics to Silicon Valley’s most self-destructive habits. In 2025, he received the Royal Society’s prestigious Faraday Prize for communicating science to the public.


Big Tech is stuck in a game it can't win
Here’s the uncomfortable truth that Wooldridge lays out: the reason companies like OpenAI, Google, and Microsoft are all racing to build AI is not because they really think it’s safe to go full speed ahead. It’s driven by fear of missing out.

This is a classic game theory problem known as the prisoner's dilemma. Two players would be better off cooperating and slowing down, but neither can be sure that the other will cooperate, so both rush ahead, a worse outcome for all. In the global race to develop artificial general intelligence, competing actors are caught in a situation where individually rational decisions have collectively dangerous consequences, a dynamic playing out in real time, according to a report by the RAND Corporation.

Wooldridge puts it bluntly: “We've got a small number of very, very wealthy companies that are busy pursuing AI, while at the same time saying that they are afraid that something's going to go horribly wrong with it. So why are they busy pursuing it? Because they think if we back down and we don't pursue it, somebody else will.”

The result is an industry that devours a lot of resources, energy, computing power, and talent, but provides no commensurate benefit to ordinary people.

Your data is fuel, and it’s running out
Data is one of the more overlooked bottlenecks in the AI gold rush. The English Wikipedia was just 3% of the training data of GPT-3. So where’s the next wave of training data going to come from?

The answer, Wooldridge warns, could be you, whether you have consented or not. He paints a picture of the near future where online influencers and ordinary users sign away their lived experiences, everything they see, say, and do, to feed AI systems essentially. Health records can be particularly helpful. He notes that the NHS in the UK and the US have huge stores of highly personal health data that corporations would pay extraordinary amounts of money for.

That is a reasonable concern. In fact, the IAPP Privacy and Consumer Trust Report 2023 found that 68% of consumers around the world are somewhat or very concerned about their online privacy, and 57% said that AI poses a major threat to their personal privacy. Millennials and Gen Z, who've grown up sharing their lives digitally, have the most to lose.

The AI that actually helps you isn’t making headlines
Here’s the irony. Wooldridge keeps pounding home the message that the AI that could actually improve your life rarely gets any press.

The talk is all about the big language models, the ChatGPTs and Geminis of the world, but behind the scenes, quietly, researchers are building tools that may matter much more. He mentions a team at Oxford that is developing an AI that can read a heart scan taken by a simple ultrasound device plugged into a smartphone and sent to a primary care physician, bringing expensive diagnostics to the point of nearly no cost.

What game theory really tells us about all of this
Wooldridge’s larger point is that strategic thinking, like the kind game theory provides, reveals patterns that we might otherwise miss. A lot of the worst behavior we see in tech and politics is a function of the zero-sum mindset, the idea that your gain has to be someone else's loss, and it also has a way of making people miserable, he adds.

Game theory tells us the better move is to search for situations where cooperation beats competition. What advice does he have for those trying to make sense of an AI-saturated world? Do not panic. Ask smarter questions, and study something you love. The future is too hard to predict to optimize your education for what jobs will be spared by AI.

He says, “I didn't get into computing because I thought it was going to give me a good job. I got into it because I was just really interested in it.”

Maybe that’s the most important lesson of all.