Digital Transformation » Systems & Software » How AI may make Black Swan events a thing of the past

How AI may make Black Swan events a thing of the past

Laurie Miles, director of analytics at SAS UK & Ireland, looks at whether it’s possible to predict and prevent the next catastrophe

History is peppered with black swan events but they’re particularly perplexing in the modern age.

Why, when we have masses of data at our fingertips, do we still get caught out by them, and are they happening more frequently?

From the pandemic to the invasion of Ukraine, the world has certainly witnessed its fair share of black swan events over the past two years. Amplifying them further is social media and a relentless news cycle, with analysis and commentary that makes us wonder whether, looking back, they could have been predicted.

Yet black swan events are, by definition, impossible to anticipate. Even if there are signs that a crisis is about to erupt, knowing when and how severe it will be is another matter.

Building operational resilience

Without a crystal ball to the future, the next line of defence for organisations is building operational resilience – with a plan in place to deal with whatever shocks come along.

During the pandemic, business continuity and disaster recovery plans that nobody had expected to put into action were swiftly dusted off to facilitate a mass move to remote working.

Not all plans were equal though. Whereas some firms sent employees home without so much as a laptop, others were agile in response to the unfolding crisis, communicating clearly and equipping people with the right tech to do their jobs.

These firms demonstrated resilience – they didn’t know a global pandemic would come along but their response ensured business as usual and sometimes unexpected success.

One of the hallmarks of a resilient organisation today is access to fast, agile cloud analytics capabilities. Real-time insights enable CFOs and other senior leaders to make fast and effective decisions, so they can channel their resources into the right areas in a timely way and limit the damage.

Organisations can develop resilience in a number of ways. Often, they’ll lean on their past experience – the historical data they hold to determine where their vulnerabilities lie. But this only provides a partial solution – as we’ve seen with recent events, history does not always repeat itself.

Mitigating future crises

Businesses can also build resilience in the moment, drawing on their currently-used metrics – sales, demand and so on – to identify potential challenges on the horizon. Every digital process produces data – including unstructured data such as text and video – which can be fed into AI models with machine learning algorithms. Modern compute power enables algorithms to interrogate large volumes of all types of data at incredible speed, looking for patterns and anomalies that people typically would miss.

When historic and current data is fed into the model, it can identify similarities between, say past market crashes and potential future ones, and trigger an alert. This predictive element gives decision-makers time to action their plans or at least keep a close eye on developments. AI and other methods, like predictive analytics, also support highly accurate scenario analysis, which can be used to assess multiple possible events to inform business continuity and disaster recovery plans.

In the financial services sector, regulations require firms to protect against unforeseen events – for example, the stress testing requirements imposed on banks. But, as we’ve seen, it’s important for businesses in all sectors to be able to make sense of, and react to, extreme events.

It’s not only the once-in-a-generation challenges this technology can tackle though. When data is analysed continuously using AI and machine learning, longer-term trends – such as shifts in customer behaviour or risk from climate change – can also emerge and indicate whether a business is ready to meet new market demands.

The one thing we can say for sure is that another black swan event will come along at some point, even if we don’t know when or what it will look like. So while advanced analytics and AI cannot make these events a thing of the past, they can certainly help organisations to be far better prepared to weather the storm.

Fortunately, many organisations are aware of the need to be insight-driven and modern technology available to them is improving all the time. AI and predictive analytics are already delivering insights in the finance sector and beyond, helping firms to build resilience and mitigate the impact of future crises.

Was this article helpful?

Comments are closed.

Subscribe to get your daily business insights