Strategic foresight is crucial to a positive future with AI

Three characteristics of this approach will help businesses, governments and social organizations anticipate and shape what’s coming: a systemic approach; openness to alternative future scenarios; and curiosity about manifold consequences.

Long before COVID-19 started turning our world upside down, epidemiologists had been warning us that it was only a matter of time before we experienced a pandemic of this nature. We knew it was coming. And we had ample experiences during which we could have honed our global response skills—among them, the 2002-04 SARS outbreak in southern China, and the 2009 swine flu pandemic.

Still, the pandemic caught the overwhelming majority of governments and businesses woefully unprepared, either because they were not paying enough attention to scientists’ alerts and the increasingly stronger signals or because they were addressing other pressing priorities.

There is a lesson here—one that we can’t afford to ignore. Leaders in businesses, governments and social organizations need to think differently about possible, plausible and probable developments of all kinds that have the potential to profoundly disrupt our societies and economies. One such development? The increasingly widespread adoption of Artificial Intelligence (AI) technologies. AI has already been significantly disruptive, but what we have seen in recent years is likely the tip of the iceberg compared to what lies in the coming years.

That’s why the world’s corporate, industry, government, and societal leaders need to bring strategic foresight to bear on AI, as soon as possible.

AI effects and potential

Consider AI’s convergence and combination with other Fourth-Industrial-Revolution (4IR) technologies. The mix of AI technologies and blockchain is fueling revolutionary smart city technologies, as evidenced by Ping An’s work in China. This as Professor Alán Aspuru Guzik at the University of Toronto Matter Lab is combining machine learning with technologies such as quantum computing to accelerate the discovery of materials that can be used, for example, for the development of new drugs.

These and other applications have great potential to help build better and more sustainable societies. Take the partnership between the UN and New York-based startup Remesh to develop an AI-powered platform that will underpin mass online conversations seeking to find out what people living on the ground in conflict zones want from peace agreements. The idea is to increase the effectiveness of peacebuilding processes—the UN estimates that two-thirds of the over 27 billion dollars spent annually on peacebuilding initiatives currently fail to lead to durable conflict resolution.

AI development could just as easily wreak havoc as support breathtaking advancements for the betterment of the earth and its inhabitants.

But AI development could just as easily wreak havoc as support breathtaking advancements for the betterment of the earth and its inhabitants. We’ve already seen it, in bias against minorities caused by unrepresentative databases used to train algorithms and in the use of “deep fakes” to fuel misinformation. What of the potential use of technologies such as facial recognition for mass surveillance, the inscrutability of decisions made by “black boxes”, and the strategic and moral dilemmas involved in AI-powered warfare?

One way or another, as Brett Winton, Director of Research at ARK Invest argues, future historians will look at our time and identify the growing maturity and use of AI, together with DNA sequencing, robotics, energy storage and blockchain technology, “and they will say: from that point, everything changed.”

How strategic foresight will help us move towards the most positive future

We need to prepare for different scenarios. We need to develop substantive plans for both leveraging AI’s potential to do good and for minimizing and mitigating its potential negative impacts—on a global scale. And if Winton and other authoritative voices, such as leading futurist Gerd Leonhard are correct, we have a limited window of opportunity in which to do so. It won’t stay open for long.

Strategic foresight is therefore essential. This discipline’s practitioners don’t claim to accomplish the impossible task of predicting the future. And admittedly, strategic foresight cannot eliminate risk or uncertainty. Nothing can do that. But it can equip us with a sound framework supported by robust methodologies to make the best decisions today, so we maximize the chances of bringing about positive outcomes in the future.

Three key characteristics of strategic foresight are crucial to this end: its systemic approach; its openness to alternative futures; and its inherent curiosity about manifold consequences. 

 1.  A systemic approach

 Most organizations today solve their problems using reductionist methods, which focus on the parts, rather than on the whole. This approach, as researchers Allenna Leonard and Stafford Beer have observed, is highly effective when causality is linear and the observer is, to a large degree, objective vis-à-vis the situation. We owe it a huge debt of gratitude, not least as it is at the heart of myriad scientific advances that have vastly improved life on Earth in recent centuries – most recently, the development of COVID-19 vaccines in record time.

Strategic foresight practitioners will never look at the future role of AI in our world through the lens of technology alone.

However, the reductionist approach is poorly placed to help us understand AI’s impact on society. Making sense of how a situation that is highly complex and plagued by uncertainty could develop calls for considering all aspects that may impact and be impacted by it. That is why futurists would never look at AI’s future role in our world through the lenses of technology alone. They take a systemic approach, and consider the social, economic, environmental and political variables involved as well.

 A strategic foresight analysis of the impact of AI on the world of work, for example, will not spend time on individual cases of companies or industries that might adopt AI solutions, or on individual tasks or jobs that might be created or cease to exist because of these technologies. Rather, it will invite us to think about how the many different aspects related to the world of work as a whole – from employment levels and social policy to pension systems and worker skills – could look like as AI adoption becomes increasingly widespread.

In doing so, it will enable us to ask more relevant questions at each step in AI’s advancement, and better steer our economies and societies towards positive outcomes.

2.  Openness to alternative futures

 Most organizations plan for the future using clear, elegant timelines divided in months, quarters and/or years. Such timelines help them plan their next steps and produce a reassuring (if sometimes illusive) sense of control. However, if we use timelines to look at the long term, they may blind us to unexpected developments, and hence leave us unprepared for them.  

This is especially the case in a world where discontinuous change is the rule. Futurist Amy Webb has noted that “[t]eams that rely on traditional linear timelines get caught in a cycle of tactical responses to what feels like constant change being foisted upon them from outside forces. Over time, those tactical responses — which take significant internal alignment and effort — drain the organization’s resources and make them vulnerable to disruption.”

 Futurists’ answer is to employ a cone that enables them to visualize plural, alternative futures from the present point.

Identifying different alternative futures helps organizations (as well as individuals) understand what their preferred future is; this “outcome” is placed on the cone alongside probable, plausible, and possible futures. Once we have defined our preferred future, we can then engage in backcasting -- in other words, we start from that future and move backward step by step to identify the necessary initiatives to be undertaken from today on, to maximize our chances of reaching it. 

Identifying different alternative futures for AI liberates us from both naïve utopia and paralyzing dystopia.

With AI, this approach will liberate us from both naïve utopia and paralyzing dystopia. The awareness that we have alternative futures and that none is guaranteed is a call to work toward building the one that benefits us the most – as businesses, as communities and as a species.

 3. Curiosity about manifold consequences

 A corollary of strategic foresight’s embrace of a systemic approach and plural futures is that it offers its practitioners tools to inquire into second- and third-degree consequences of the phenomena under investigation. Different methodologies from the discipline’s vast toolkit actively force practitioners to consider the knock-on effects of possible future developments.

 For example, in the Futures Wheel methodology futurists depart from a central trend or event from which they derive different first-order impacts. Each of those impacts has second-order impacts, which may, in turn, also have third-order impacts.

 To picture the practical implications of next-order impact considerations, take the case of a scenarios exercise in which author Stephan Talty confronts readers with a series of fundamental questions, such as: What if find ourselves living with AI that both is able to perform multiple tasks and has a cognitive capacity that far exceeds ours? In one 2065 scenario, Talty discusses whether in those circumstances the judiciary will consider AI to have conscience or personhood. If so, whether AI will be granted control of its own money and, in that case, how it would use it, and with what implications for humans.

Building and refining our best approach

The benefits of strategic foresight have been proven. In the early 1970s, Shell executive Pierre Wack developed a scenario in which OPEC countries would restrict supply for political reasons. The company prepared for price rises by adopting shorter-term contracts. The result: When the cartel announced its embargo in 1973, Shell was able to boost to its margins much faster than its competitors. More recently, researchers René Rohrbeck and Menes Etingue Kum conducted a study of 83 European multinationals that showed that superior foresight practice in 2008 was a powerful predictor of being an industry outperformer in terms of both profitability and market capitalization growth in 2015.

A growing number of companies, governments and international organizations are embracing strategic foresight.

And a growing number of companies, from Cisco to L’Oréal and Daimler to Pepsi, understand the many benefits strategic foresight brings. Governments are also increasingly embracing strategic foresight as a way to improve the quality of their polices. In countries such as Finland and Singapore, groups working on strategic foresight are part of the Prime Minister’s Office. The United Arab Emirates has a Ministry of Cabinet Affairs tasked with “developing a strategy that ensures all sectors’ readiness for the future’s variabilities”. International organizations, such as UNESCO and the OECD, also use and promote the wider adoption of strategic foresight.

It's time to turn that lens on AI. As AI combines with other 4IR technologies to help set in motion profound changes at breakneck pace in the coming decades, strategic foresight offers leaders in businesses, governments and social organizations invaluable tools to help us build and constantly refine a human-centric approach to this new stage in our social and economic lives.