Predicting, particularly when it is about the future, is difficult!

The joy of scenarios

At a recent conference we shared the observation of Chris Caren CEO of Turnitin LLC (an Internet-based similarity detection service used by Academia), that "Most of our employees are [software] engineers, we have a few hundred of them, and I think that in 18 months we will need only 20% of that number and we will be able to hire them directly out of high school rather than from 4-year college [programs]”. It was one of a number of cues that we have picked up on that suggest a ‘direction of travel’ for AI enabled job displacement.

What was interesting was that during the Q&A an audience member suggested that Chris Caren’s observation was a prediction that would no doubt be wrong (because all predictions are wrong) and so any inference we might make from this would – to borrow a legal metaphor – be ‘fruit of the poisonous tree’. In this our commentator was perhaps reflecting the thoughts of US economist Edgar Fiedler who once noted that: “He who lives by the crystal ball soon learns to eat ground glass”. And he is not wrong – however we are not in the business of making predictions.

We agree with Nobel Laureate Niels Bohr that “Predicting, particularly when it is about the future, is difficult”, but that does not mean that we are without tools that allow us to prepare for what is to come. The question we face is: can we imagine this future so that we may evolve to meet it, or are we simply to face what comes and hope for the best? We are scenario planners, and in this post we thought it might be helpful to reflect on scenario planning and what a useful tool it is when faced with future uncertainty of the type AI is creating.

Scenario planning emerged more than 60 years ago in response to the challenge of decision-making under unpredictable uncertainty. Futurist Herman Kahn, one of the very early exponents of scenario planning was described by Scientific American as: “thinking the unthinkable”. Shell was an early pioneer, using this technique in its much-heralded Group Planning department under Pierre Wack. Wack saw scenario planning as a way of “changing the mental map of managers” i.e., getting them to consider alternative futures than the ones they were currently aligned with. Peter Schwartz, US futurist and co-founder of the Global Business Network which specialized in scenario planning, believed that scenarios were tools for ordering one's perceptions about, and potential responses to, alternative future environments, in which one's decisions might be played out.

Scenario planning is a thinking tool for use in strategic conversations; in short, a technique for challenging assumptions and pre-existing models, and encouraging fierce strategic dialogues around how best to see opportunities and navigate an uncertain future (Van der Heijden - Scenarios: The Art of Strategic Conversation). Thus, the key is to think of scenarios not as predictions about the future, but more as a way of strategically reframing what is going on. It is a technique for outlining possible  “future histories”. This mapping out of scenarios allows us to reflect and decide what the best possible strategic place in which to play in the future is, given our talents. To reiterate, we are not making predictions. We are instead identifying key evolutionary pathways and critical uncertainties, providing an informed view on how they might evolve and how to prepare for these different eventualities in a timely way. It’s about creating resilience.

The approach we use is based on an adaptation of the Ramirez and Wilkinson technique, as outlined in their 2018 book Strategic Reframing, which introduces the concept of there being “three arrows of time”.

First, there are things from the past catching us up now or later (the momentum of the past); things that have already happened that will continue to have an impact in the future. We describe Ramirez and Wilkinson's “first arrow of time” as “power trends”. The key here is to identify the fundamental big-ticket issues that are most likely to affect your industry. You are looking for trends that are presently widely acknowledged, extensively shared and debated by influential sources. Moreover, there is sufficient evidence to suggest that this is a substantive trend rather than a short-term fad that's likely to disappear or correct itself. These themes reflect the core underlying drivers of what is happening in the world.

One of the most obvious trends that it would be hard to ignore is the growth in available data and the extent to which this is structured or unstructured. As more of our lives are captured in data, we can expect the reach of AI to continue to spread.

The second arrow represents the present going forward—the future embodiment of action planning. We think of this second arrow as “accelerating present”. This talks to the likely future impact of decisions taken today. Here we embrace William Gibson's point that the “future is already here, it's just not evenly distributed” (2003). You have an opportunity to look at particular features and developments of what's happening at the leading edge of thinking as it relates to your industry. This will help you to better understand and start acquiring a more informed read of how this is likely to play out in the future. Bear in mind the trap of the Amara's Law. This tells us that there's a tendency to overstate the impact of specific technical developments and issues in the short-term, whilst at the same time still managing to underestimate their power and impact a decade or so out.

Here we might consider signals coming from funding data, patents, earnings calls etc. or developments in AI chip technology or indeed more qualitatively the views of CEOs when they muse on the likely size and shape of their future workforce. There are any number of cues that we may consider.

This takes us to Ramirez and Wilkinson's third arrow: “the future coming towards the present” — future developments coming towards us independent of our will. You must identify and prioritize the critical uncertainties heading your way; by “critical” we mean the uncertainties that are out there are most likely to have a major impact on your industry.

Again, it is not about predicting what is likely to happen, but more to do with using our analytical and creative thinking skills to pinpoint those uncertainties that have a high probability of happening and are also likely to make a high impact on your industry. We can then compare and contrast these with scenarios where there is a lower probability of this uncertainty happening, or less likely to make an impact on your industry, whilst also looking at all points in between.

Identifying critical uncertainties is the biggest challenge – particularly in an AI landscape that is characterized by discontinuous change (which may in itself be one of the critical uncertainties). However there is an abundance of commentary by the informed (and not so informed) that provides source material – your job is to evaluate this, adding your own creativity and categorizing future uncertainties through the lenses of degree (of uncertainty) and potential impact.

The point of convergence of our three arrows is where we consider a range a plausible, possible futures – where we synthesize what we have learnt from our three arrows into key dimensions which can then frame future scenarios. And then with each of the scenarios, we can consider implications and potential responses.

Its not about predicting which scenario will come to pass, but rather challenging assumptions, fostering a fierce strategic future-focused dialogue and building resilience.