Many people think Sapiens: A Brief History of Humankind by Yuval Noah Harari the single most important book they have read, and a nonagenarian friend said it made him see the world in a new way. The book has been translated into 40 languages, but the commonest question Harari was asked in his interviews about the book was “What next?” His new book Homo Deus: A Brief History of Tomorrow is an attempt to answer that ultimately unanswerable question, and he talked about it in London last night in the first performance of a world tour.

Asked to describe the central thesis of the book, he began by saying that throughout human history we have battled against famine, plague, and war. We prayed to gods and tried many political systems but got nowhere. But in the last century developments in science, technology, and medicine mean that we have made substantial progress in countering these evils. More people (three million) now die from overeating than from starvation (one million). People are more likely to die of old age than from an infection. Largely because of nuclear weapons, he argues, the past 80 years have been the most peaceful in human history: more people die from suicide than die from war or terrorist attacks. Famine, plague, and war are by no means gone, but technology has pushed them back for the first time.

So what next? The big human projects, he said, are to overcome old age and death, find the key to happiness, and become gods. He emphasised that he meant become gods literally. Humans are close to achieving divine abilities: overcoming death to achieve immortality; creating and designing life; and perfecting bodies, brains, and minds. Indeed, humans may exceed gods, who created only organic beings, by creating inorganic beings through artificial intelligence.

Humans have turned their attention from outside issues like famine and war to re-engineering and manufacturing brains and bodies. Biology and computing have to come together to do this because “the human brain is not adequate.” The result, Harari suggests, could be “the end of humanity” because artificial intelligence will create something much more powerful than humans. Humans get tired, make mistakes, fail to think rationally, and forget, whereas machines do not.

Harari sees “as a distant possibility” the rise of a superclass of humans who control these technologies and a “useless class” who are no longer needed because machines can do their jobs better than they can, no matter whether they are taxi drivers, chefs, GPs, editors, brain surgeons, or rocket scientists.  Harari compared this with the industrial revolution, which created be proletariat. Machines have since then steadily replaced physical work, leaving humans to move to mental work, but now machines are replacing that too. It’s already happening with armies, which once consisted of millions of men but now are increasingly dominated by small groups of “super- warriors” who control technologies like drones and fight cyber wars.

The divide between the superclass and the useless class is likely to be written into biology with the superclass having much superior physical and cognitive capacity and living much longer. The two classes might evolve into different species.

People are less likely to be destroyed by machines than to be pushed to one side, argues Harari. But to be irrelevant and useless may be worse than being destroyed.

Neoliberalism, Harari believes, leads directly to the domination of machines and the algorithms that inform them. The essence of neoliberalism is that “markets, made up of millions of decisions combined, know best.” Governments, he suggests, like neoliberalism because they don’t understand what’s happening in the world—nobody does. Stock exchanges might be seen as the best decision makers in the world, but, argues Harari, algorithms built by artificial intelligence using big data will soon be better. Machines will take over from the market.

But doesn’t that mean that we, the useless class, will be able to play football, travel, read poetry, and discuss philosophy? Beware, answers Harari, because once you lose economic power you lose political power. You are dispensable. Then happiness in life depends on feeling your life has meaning. Where is the meaning in a life of leisure? And humans are hard to satisfy: people who have written great symphonies want to write more.

Might you be able to leave us on an optimistic note, asked a questioner. Nuclear weapons, responded Harari, provide an optimistic example. Chekhov famously said that if a gun is seen in the first act it will be sure to go off in the third, and many clever people were convinced in the 50s and 60s that there would soon be a nuclear war. But there hasn’t been, and nuclear war seems less likely now than 60 years ago. Humans can manage to live with—and even benefit from—powerful technologies that have the potential to destroy them.

Many other interesting ideas emerged during the 90 minute performance of Harari, and I’ve summarised some of these on my own blog. The topics include philosophical problems becoming practical problems; elections not including what matters most; how we are selling our personal information for baubles as Native Americans sold Manhattan for beads; how decisions like Brexit are made now on feelings but might soon be made based on algorithms; how electronic books read us better than we read them; and how machines might be better therapists than human beings. I’ve also on my website included many quotes from Sapiens.

Competing interest: RS was an unpaid trustee of C3 until last year.

Richard Smith was the editor of The BMJ until 2004.