Enter the Age of With: Where humans work side-by-side with machines

Deloitte Analytics' Catherine Stretton: The time is right for AI implementation.

According to research firm Gartner, by 2021 artificial intelligence (AI) will create $2.9 trillion of business value and 6.2 billion hours of worker productivity. AI is essentially a computer or software system that uses algorithms to make it possible for machines to learn from experience, adjust to new inputs and perform or simulate human-like behaviour or tasks.

For a number of years, companies have targeted low-value opportunities for task-based automation, but increasingly seek to incorporate more advanced analytical and AI technologies.

Catherine Stretton, partner at Deloitte Analytics, believes the time is right for AI implementation. Corporate interest has moved beyond conference discussions to tangible business cases. Catherine comments: “In 2016 the market wasn’t ready, but interest has slowly built over the years. Clients are interested in the beautiful problem-solving abilities of AI. We seldom talk about the technical details of the technologies themselves, but rather the solutions they offer.”

When applied to finance, digital technologies allow CFOs to manage traditional responsibilities like internal controls, compliance, and closing the books faster and with more efficiency. For the finance team, less time is spent on routine tasks and more focus on understanding the levers of revenue growth.

There is an obvious business case for AI in risk management. Here, these technologies are adept at spotting patterns in historical data. AI can be used to detect fraud, uncover money-laundering or pinpoint customers with high credit risk profiles. But combing through data is not where it ends. Technology is also able to learn and identify new methods in hacking or cyber-attacks.

For Catherine, one of the biggest competitive advantages that AI offers is to slash the time needed in forecasting and financial modelling. This is bringing forecasting ever closer to real-time. In the mining industry, for example, it used to take weeks to do a production forecast. Today, technology makes this possible in a few hours.

Compelled to compete
While the business benefits of AI are abundant, locally the corporate market adoption of AI has lagged. Catherine points to hesitations from executives, who resist AI due to perceived costs and feared complexity. They may be already grappling with analytics and large-scale data projects. For Catherine, this is the perfect time to layer on AI. In contrast with other technology projects, AI implementation can be less time-consuming and demanding than expected. She comments: “Executives are often surprised at how lightweight an AI project can be.”

In general, some local indications are positive. Catherine says that our universities are grappling with the issues surfaced by AI and producing strong graduates. There are also a number of exciting newer AI entrants. Cape Town based start-up DataProphet, a leader in AI for manufacturing, was included as one of World Economic Forum’s 56 emerging tech firms for its 2019 cohort of Technology Pioneers. Brandseye is another example of the ‘Age of With’ by combining AI and human decision-making to monitor online conversations and rank these according to value and risk.

In the banking space, Catherine believes that our legacy banks will be compelled to use AI to compete with the newer entrants. Here, the incumbents have several highly customised systems, computing environments and processes that limit their ability to manoeuvre. This is a stark contrast to Tyme Bank, whose core technologies are cloud-hosted. Likewise, Discovery Bank’s capabilities to calculate individualised interest rates based on customer behaviour are made possible through advanced technologies.

Customers will be attracted to digitally astute banking services that offer a more personal experience. According to Gartner, the real commercial value in AI is how it enhances the customer experience – reducing mistakes and offering personalised service at scale.

An example of a more intelligent customer experience is Deloitte’s work in the UK developing a technology that can ‘listen’ to call centre conversations. The technology can pinpoint moments where the voice is stressed as well as indicate real enthusiasm. This allows call centre management to respond to instances where the customer may have felt wrongly pressured to buy or follow-up on hot customer leads.
With power comes responsibility

As AI becomes more powerful, many more groups are becoming interested in its responsible use. The ethics of data harvesting are increasingly under the spotlight. Another potential issue is bias based on race, gender, age or location, and bias based on a specific structure of data, have been long-standing risks in training AI models. 

Gartner predicts that 75 percent of large organisations will hire AI behaviour forensic experts, privacy and customer trust specialists to reduce brand and reputation risk by 2023. Organisations such as Facebook, Google, Bank of America and NASA are hiring or have appointed AI behaviour forensic specialists who primarily focus on uncovering undesired bias in AI models before they are deployed.

South Africa, with our high unemployment figures, has specific ethical challenges with regards to AI adoption. Catherine believes that deciding on the human skills needed in a workplace filled with AI solutions is hard but necessary work. “It may be possible to deploy technology models at scale, but should you? Leaders need to determine their AI strategy, including how to reskill employees and how to communicate this change within their organisations.”

New technologies will not mean that human intelligence is under-valued. In the ‘Age of With’ a world where people work side-by-side with machines could enhance human motivation and satisfaction. The need for human involvement is not going away, and the value of powerful, esoteric capabilities like imagination cannot be underestimated. Also important is the human capability for empathy and complex problem-solving.

What is South Africa’s AI strategy?
In 2017, the Canadian government kicked off a $125 million AI strategy, the world’s first national strategy. Other countries are following their example, including identifying a unique AI positioning.

“South Africa needs a national AI strategy and positioning. With our major societal challenges, including in the areas of healthcare and education, our positioning could potentially be centred on building AI for good,” says Catherine.

Our Communications Minister Stella Ndabeni-Abrahams kicked of the first 4IR Commission meeting in May 2019 and has committed to producing a 4IR strategy document by March this year.

AI has incredible potential in healthcare. An international example of AI for good, Deloitte and the Tisch MS Research Center of New York are using data science, AI and machine learning in a research project to find patterns that may relate to the cause of multiple sclerosis (MS). In just two weeks, a Deloitte data scientist and the Deloitte US innovation team were able to identify molecules likely to be correlated to MS. This research, when done by humans, may have taken up to 10 years.

Deloitte’s AI capabilities
Globally Deloitte has invested heavily in AI – in specialists, capabilities and assets. The firm has a dedicated AI studio in Frankfurt that advises Deloitte offices throughout the world. In South Africa, Deloitte has a 30-member strong AI team.

“We are a globally connected team that is able to share business cases across the network. Our biggest advantage is that our AI team collaborate with Deloitte’s many industry experts. We are close to the burning problems to that our clients in different industries face. We understand the nuances. This means we are better able to identity technology solutions to business problems,” says Catherine.

A CA by training, Catherine has been with Deloitte for 24 years, both in the firm’s UK and South African operations. She has spent the last seven of these in Deloitte analytics division and also leads the firm’s analytics team in financial services. Much of her work involves helping financial services clients to manage and interrogate their data to enhance business performance. For Catherine, the best part of working at Deloitte is the company’s entrepreneurial nature and how it’s possible to reinvent a career a number of times.