Collaborative intelligence: humans & AI joining forces to support data-driven decision-making

Collaborative intelligence: humans and AI joining forces

In the early 19th century, textile workers in Nottingham rebelled against their factory owners

As factory owners began to use new machinery that reduced the number of employees and factories they needed, workers felt that their skillset was being wasted and their livelihoods threatened.

This rebellion was the Luddite movement. The term ‘Luddite’ has since been used to describe those who opposed industrialisation, automation, and in more recent times some cutting-edge technologies threatening to disrupt the mainstream.

When it comes to artificial intelligence (AI), you can sympathise with the Luddite philosophy to an extent. The idea that we can teach machines to think for themselves is an Academy Award-winning one, from Ex Machina to I, Robot. If a machine can do your job, why are you needed?

Fortunately, that level of AI is a pipe dream. We don’t get the most out of AI by using it to replace employees, but by using it to augment existing roles. It’s an effective tool for enabling employees to focus on higher value, more strategic tasks, while AI takes on the high-volume heavy lifting.

Humans and machines work well together as they actively enhance each other’s complementary strengths: the leadership, teamwork, creativity, and social skills of the former, and the speed, scalability, and quantitative capabilities of the latter.

This synergistic approach – the concept of ‘collaborative intelligence’ – is what organisations should be working towards as they shift to deeper data-driven decision-making.

The opportunity

While the concept of machines augmenting human activity has been with us for some time, a significant change in working patterns combined with increased pressure to support faster and higher quality decisions when adjusting to rapidly moving situations means we expect to see further experimentation and investment in 2021 and beyond.

A simple but obvious example is in employee and customer interactions, where making the right decision can have a huge impact on the bottom line. The last 12 months has seen hundreds of millions of consumers introduced to online shopping, with businesses scrambling to digitally transform and meet that demand.

Given such a swell in customer data and the need to differentiate from competitors, the most successful businesses have used AI powered systems to provide their employees with a better understanding of the customer – from previous behaviours to purchase history. Analytics support gives them the insights and context to personalise interactions and communications so they can perform at their best.

Collaborative intelligence will also help optimise and enhance aspects of customer communications in high value areas such as fraud detection, customer service and issue handling. However, to make effective use of analytics and AI to support decision-making, organisations need the right skill sets and the right people.

Training and upskilling talent

To understand the insights that data offers us, we need to have a basic understanding of how customer data is collected, how to read the findings and what they mean at a human level.

This sort of ‘data literacy’ – being able to read and translate data from its raw form into useful insights – is hugely important to the modern business. You don’t need every member of the business to be fluent in code, but they do need the relevant skills to navigate, read, and comprehend different data sets.

However, employees with such a skill set are in short supply. Many pre-COVID businesses often had no burning need for sophisticated IT infrastructure, with fairly unsophisticated data collection strategies. The lack of data skills in their workforce reflects that, and now those businesses need to solve these challenges if they’re going to maximise the gains from their data insights.

Young people entering the workplace may seem an obvious solution, but they haven’t necessarily been given the opportunities to develop their own data skillset. Exasol’s D/NATIVES report has found that more than half (57%) of youngsters between 16-21 consider themselves data-literate, while the number of young people taking IT subjects has dropped by 40% since 2015, according to Work and Learning Institute.

That means businesses need to invest in their own solution to the problem. Forward-thinking organisations need to help their workforce understand and trust the insights that they receive from data.

In real terms, that means training, training, training. The most committed organisations establish a ‘Centre of Data Excellence’ to coordinate the education and use of data internally. At the top of the tree, the Chief Data Officer role is becoming more and more common as large organisations recognise the need for properly managed data.

A concerted effort to incorporate data into the business, and to help everyone in the organisation recognise and understand it, is the nucleus of a proper ‘data culture.’ As data begins to be built into performance KPIs and company meetings, and employees understand it more naturally, organisations will begin to see the potential of a collaborative intelligence approach.

Collaborative intelligence in action

As the insights from data and AI become more sophisticated and robust, it’s possible to automate business-critical processes from start to finish.

General Electric is using intelligent agents to empower its maintenance workers with data-driven insights as they are operating industrial equipment. The workers interact with AI agents to source information such as diagnostics and predicted failures in order to make accurate multimillion-dollar operational decisions.

This example is based upon the building blocks of collaborative intelligence: a data-literate workforce with the skills and awareness to recognise the value of data that their business has made easily accessible to them. Organisations that choose to invest in the data literacy of their workforce are the best positioned to take advantage of such opportunities.


Can Self-Supervised Learning Teach AI Systems Common Sense?