Source:
https://scmp.com/comment/opinion/article/3077176/forget-fancy-ai-predictions-world-must-come-together-think
Opinion/ Comment

Forget fancy AI predictions. The world must come together to think for ourselves in this coronavirus pandemic

  • In such extraordinary circumstances, forecasts and predictions by clever computers or AI-driven algorithms cannot provide specific solutions. These can only come about by putting our smartest thinkers together through multilateral action and global cooperation
Japan’s Prime Minister Shinzo Abe attends a video conference with other G7 leaders from at his official Tokyo residence on March 16. Photo: Reuters

Uncertainty over our future is arguably more acute today than at any point in my lifetime. Angst during World War II about the implications of a Hitler victory was undoubtedly acute. So, too, about the implications across Asia of Mao’s victory in China. The nuclear missile crisis at the height of the Cold War without doubt sent shudders across the world.

But none of these had the existential qualities of the toxic cocktail of at least three colossal uncertainties that face us today: a mounting Thucydidean conflict between hegemon United States and upstart China, the coming climate crises linked with global warming, and a pandemic that is set to test our global health systems, cause perhaps millions of deaths, and trigger a global depression.

Francis Fukuyama’s optimistic The End of History and the Last Man, published just 28 years ago, seems today as complacent and naive as it was wrong. It also illustrates our prodigious capacity to mis-predict the future.

I remember the well-respected Far Eastern Economic Review predicting in 1975 that the most prosperous countries in Asia at the end of the century would be the Philippines and Myanmar. Whoops.

And yet the futurology industry and the population of consultants worldwide who are confidently – and profitably – offering to forecast our corporate or economic futures are thriving as never before. Armed with exploding computing power and algorithms driven by artificial intelligence, many bulge with self-assurance and a newly minted sense of infallibility.

Timely, then, to stumble upon the fascinating new book by business consultant Margaret Heffernan, Uncharted: How to Map the Future Together, to bring these new “masters of the universe” back to earth.

She complains of those executives worldwide who “have come to expect the future to be minutely and perfectly predictable” – the naive fodder upon which consultant forecasters so eagerly feed. She also blames the false confidence marketed by such consultant advisers for encouraging overdevelopment of the vulnerable and fragile global supply chains on which so many multinationals so heavily depend.

I do not think all the blame can be dumped on self-serving consultants. Many CEOs I have worked with have craved a manual that can be lifted off the shelf to address and resolve any sudden crisis or challenge. They actively seek documents that predict, rather than encourage preparation. I always told them – often to no avail – that no such manual exists. If it did, there would be no nail-biting need to fret over the outcome of every weekend’s football matches or the winner of Race Four at the Sha Tin horse races.

It seems obvious to say the future is not – and never will be – perfectly knowable, or even predictably manageable, but is instead the product of a complex contest between numerous forces that are often obscure and unpredictable.

It seems very few people recognise the significance of serendipity – the potential for random and wholly unexpected events or developments to wreck even the best-laid plans. I have always believed that serendipity is among the most important words – or concepts – in the English language, even for caring parents who, in a place like Hong Kong, go to such lengths to chart a predictable future for their children.

As Heffernan notes: “Unique and rare external events may render what was formerly predictable suddenly unforeseeable”. Think the Tokyo Olympics, or the complete collapse of the world tourism industry, or, for anxious Hong Kong students, the postponement of this year’s university entrance exams.

A camel waits at the Giza pyramid complex on the southwestern outskirts of the Egyptian capital Cairo on March 13. The outbreak of the novel coronavirus has put the US$1.7 trillion global tourism industry under threat. Photo: AFP
A camel waits at the Giza pyramid complex on the southwestern outskirts of the Egyptian capital Cairo on March 13. The outbreak of the novel coronavirus has put the US$1.7 trillion global tourism industry under threat. Photo: AFP

Heffernan reserves some special spleen for the world’s hi-tech, AI-inspired companies, and the “utopian fantasy” that their predictive, algorithm-based systems spell an end to unpredictability. She mischievously quotes Cathy O’Neill, the Columbia-based mathematician and author of Weapons of Math Destruction, who argues that algorithms are no more than “opinions encoded in numbers”, and Philip Tetlock at the University of Pennsylvania, who has famously argued that the more famous forecasters are, the more likely they are to be wrong. Remember Fukuyama.

The kind of forecasts she respects are those linked to global warming or pandemics or global stock market crashes – forecasts that are “generally certain, but specifically ambiguous”: we know for sure they will occur and recur but never commit to saying exactly where or when.

She insists that any leader worth his or her salt – whether in business or in government – must never put their trust in machines, no matter how cleverly contrived. Instead, they would be “prepared to navigate the unknown in pursuit of the ill-defined because they realised that the only way to know the future is to make it.”

And so to the unprecedented set of existential challenges in today’s perfect storm. It is in such extraordinary circumstances that computer models or rule books have to be put aside, enabling us to think for ourselves.

Better still, as Richard Hatchett, CEO of the US Coalition for Epidemic Preparedness and Innovation, insists, we should concentrate on being prepared rather than making predictions. We should also, above all else, “own the pandemic problem collectively and find a collective solution” globally: “We must participate in a global ecosystem of epidemic responsiveness.”

Focusing on the “generally certain”, we need to assume the pandemic will be with us for a year, maybe a lot longer, with no reliable vaccines available until well into next year. Governments and businesses have to work though how to respond to these general certainties. They will do this more effectively if they work together, rather than alone.

Setting specific targets, like getting back to business by Easter Sunday, or reopening schools on a specific date, is both naive and foolish.

We can use models and fancy AI-driven algorithms if we like, but Heffernan’s advice would be to use them as a means of exploration, not as engines for specific solutions. At the end of the day, it will be smart people working smartly together across the world that will lift us out of this dreadful hole most quickly and effectively. And if that is not living proof that globalisation and multilateral cooperation give us superior results, I do not know what is.

David Dodwell researches and writes about global, regional and Hong Kong challenges from a Hong Kong point of view