The Business Courses Navigating the Future of AI
How do leaders navigate the most exciting, disruptive and risk-loaded technology of our time? Business schools are going all-out to help build essential skills and knowledge in AI – fast.
As any technologist will tell you, artificial intelligence is not a new beast. The march of intelligent machines has long been a sci-fi trope and the term as it relates to computing was coined by researchers in the 1950s. But the launch of OpenAI’s ChatGPT large language model (LLM) in November last year has been a shrill wake-up call for leaders: AI is your job now, too.
“It might not be new from a technical perspective but generative AI is revolutionary,” says Judy Slatyer, head of the Responsible AI Think Tank at Australia’s National AI Centre. “Within two months, 100 million people were using it – that’s what we’ve got to get our heads around.”
Slatyer, a leader with global experience, is entrepreneur-inresidence at CSIRO’s Data61, which houses the National AI Centre. While she’s quick to point out she’s not a technologist, Slatyer brings real-life experience on how companies need to adapt to technologies. An advocate for continuous learning, in 2022 she took a six-week online executive education course to build her knowledge about AI’s implications for business strategy.
“It’s important that business schools and educators help leaders ask the right questions and then walk alongside them as they get the answers.” As generative AI goes deeper into companies and supply chains, she adds, leaders need to understand the complex opportunities – and the risks. “AI has very long tentacles and many different uses. Educators need to help CEOs and others to make sure they explore all pieces of the puzzle, including the impact.”
Aided by a highly educated population, Australia is well-placed to capitalise on the latest AI revolution but Slatyer cautions that the most recent Responsible AI Index shows a drop from 8 per cent to just 3 per cent of businesses reporting that they rate themselves as “mature” in the space. “We need good frameworks to help CEOs explain how they’re using AI in their organisation – to take accountability for the strategy and to overcome their fears about the risks,” she says, adding that few executives can confidently do that today. Business schools are developing a range of courses designed to help leaders build their AI muscles. Leading exponents from five schools share the key capabilities they’re targeting.
1. Recognise the power of generative AI
Within 48 hours of ChatGPT’s public release in November 2022, Professor Mary-Anne Williams hosted a pop-up workshop on generative AI for the UNSW Business School, where she is director of the Business AI Lab. “It is so exciting,” she says. Williams, who did her PhD in AI, has been working in the field for about 30 years and in generative AI since 2017. “It’s not about copying and pasting – it’s about using AI to add value and adapt it to a particular need or circumstance. That’s the human part.”
She warns that we should strap in for upheaval. “This is a massive step change – this turns everything upside down and we have to remake the world. Generative AI is very powerful and if we harness it in a safe, inclusive, responsible way, we can do something quite extraordinary. As educators, we are at the coalface of guiding that.”
There are fears – particularly in universities – about generative AI’s leg-up for copying but Williams flips that. “Copying is important for innovation. If we’re not producing people in Australia who can adapt what’s already out there, we’re creating generations of people who think they have to reinvent the wheel – we’ll never get ahead in technology. Of course, we have to ensure students don’t plagiarise but if we’re not encouraging them to use AI to explore, amplify and augment what they can do and how they can do it, that’s a real problem.”
As Williams points out, calculators and even shovels were once revolutionary new tools. “The surprise is the speed with which it’s happening and also the magnitude of the change. Nobody has seen anything like this – ever. We’re going from riding a horse to co-piloting a spaceship in one step, without a bicycle, car or plane in between. It’s one huge leap so there’s a lot of risk and uncertainty.”
Creating value from AI is something the UNSW Business AI Lab, which launched last October, is focusing on. “It’s not about profit, it’s about value,” says Williams. “Which AI should we be implementing? Which will have the biggest impact, the most ROI and help our customers create more value for themselves? We talk about value because our definition of business is every human endeavour – including social enterprise, government and for-profit businesses.”
Associate Professor Babak Abedin from Macquarie Business School is equally excited about the potential of AI, beyond operational automation, to improve productivity, which many are already familiar with. “AI is going to help humanity and represent us in places and cases and applications that would either not be feasible for humans to be there or it would not be as efficient,” he says. “An extreme example is having AI represent humans under the oceans and in space to make real-time decisions.”
2. Know enough to be dangerous (in a good way)
Leaders don’t need to learn to code but they do need to develop an appreciation of what’s under the hood. “Our program has no formulas, no code at all,” says Professor Matthias Holweg from the University of Oxford’s Saïd Business School, which introduced an online short course in 2019. “We teach leaders to understand how these methods work in an illustrative way so they can go away and understand the potential and the risks.
“I tell leaders they need to think like taxi drivers who know very well what kind of car they want. Hybrid, electric or diesel? Sedan or an SUV? That brand or another? The cab driver doesn’t need to know how the fuel injection works but he or she knows very well whether she wants fuel injection or an EV.”
Executive education in AI aims to get students to the level of what Holweg calls the competent customer – or clued-up taxi driver. “You work in this space and you know what’s good and bad, what’s risky and what’s reliable. Getting leaders there means we can use AI to society’s advantage.”
“They need to understand from a business point of view where the intersection with AI is and the value you can create,” says Lamont Tang, program director and director of industry projects at the Australian Graduate School of Management (AGSM) at UNSW Business School. “We build those foundational skills so leaders can navigate between the hype and the noise.”
Part of that is knowing the right questions to ask, much of which comes down to the datasets on which AI and machine learning (ML) algorithms are trained, along with a clear view of the purpose. “When we engage with scaleups and startups,” says Tang, “we always ask what the data asset looks like, how it will be used, who will use it and other questions that leaders should be cognisant of.
3. Prioritise fairness – and be diligent about data
There’s a lot of talk about the issues around bias in AI and many infamous examples of how it gets companies into trouble, from algorithms identifying nurses as female and doctors as male to racial biases in recruitment models.
“Knowing what goes into your model is really important,” says Anita Arbogast, executive director of the Centre for Business Analytics at Melbourne Business School (MBS). Big organisations will need to understand how to feed their own AI, using data from inside their walls. “Where security and accuracy are paramount – such as in healthcare or banking – that’s really important. Large language models (LLMs) have been fed copious amounts of data from the web – not necessarily the most reliable source. Generative AI makes the potential to spread misinformation so much faster.” As with other schools, MBS is devising a variety of courses to get leaders across this risk, stressing the importance of having a clear, considered view of their data.
That also means getting a handle on whether your dataset could hold historic information that could lead AI to make biased decisions, such as discrimination in bank lending based on race or gender. Professor Zsolt Katona at Berkeley Haas business school in California prefers to talk about fairness. “Bias can be mixed up with the statistical meaning of ‘bias’,” says Katona, lead faculty for the school’s online course in AI business strategies and applications. “That’s why I prefer to use ‘unfairness’. Contrary to what some believe, a lot of human tinkering is involved to get these algorithms to work, even if they appear autonomous at the end. We need a lot of human input – and the more eyes and the more different types of eyes the better.”
Data is the life source for AI and business schools make it clear that doubling down on data privacy and security is another responsibility that goes all the way up to the CEO. “Look at the two recent major data breaches in Australia – there was information kept of people who were no longer customers,” says Arbogast. “Leaders should realise their organisations need to hold the bare minimum of data to service a customer.”
Tang says the evolution of LLMs means they will be more efficient with less data. Leaders who ensure their businesses prioritise data quality over quantity and are transparent about how they use data could win big. “Signalling that you’re taking responsible steps for all your processes and your technology is how organisations can distinguish themselves from their competitors.”
In 2019, after consultation with hundreds of stakeholders, CSIRO’s Data61 produced Australia’s AI Ethics Framework. “There are eight principles and they have much in common with frameworks in the US, Canada, the EU and China, and big tech companies,” says Abedin. “For executives, the framework is a good point to start thinking about what they mean for their company.”
4. Be able to explain your own AI
No more throwing to the tech team – all leaders, including line managers, should be able to clearly explain how they are using AI, along with their plans for their workforce as AI takes over more tasks.
Short courses are the smartest path for essential grounding. “It’s not much use coming into our executive diploma in AI unless you understand the basics,” says Oxford’s Holweg. “We get people from every sector for our six-week online program – many from government, consulting, healthcare, pharma and legal. The least represented is manufacturing – they think, ‘Oh, we make stuff, we don’t need to worry about it’, though I would beg to differ.” Holweg’s point is that AI is at the heart of industrial automation.
The academics are reassuring about the understandable anxiety experienced in the face of the AI learning curve. “We look through a business lens at what leaders – who are not coders or mathematicians – need to understand in terms of the concepts to bring to the fore and also to ensure they can ask the right questions around AI,” says Arbogast.
“Our courses are suitable for people with absolutely no STEM background,” says Berkeley’s Katona, echoing the sentiments from many of the other business schools. “We try to explain how AI works without really any equations. Our goal is to teach them language so that they can communicate about it.”
“This year, we’ve added more case studies about AI and machine learning,” says Tang. “We’re not going to teach our executives to code but we can give them a little taste of how just a few lines of code can help automate workflows or reveal new growth opportunities. In executive education courses, we’re teaching data literacy and data analytics and, this year, we’ve had a lot of incoming demand to upskill leaders and managers in generative AI.”
More help is on the way – 2024 will see multiple new courses at business schools to get leaders up to speed on generative AI. “Organisations are reaching out to us and I’m sure to others,” says Abedin. “As far as I know, there is no single course out there to cover everything but there are a lot of discussions and there’s a lot of awareness. Moving forward, you will see more systematic learning and teaching programs.”
5. Keep humans in the loop
“We want to shift the conversation from focusing on the tech because the future is all about human-AI collaboration – how we work together,” says Williams. “AI is awesome but without humans it doesn’t have direction, motivation or real problems to solve. That’s what the UNSW Business AI Lab is oriented towards, plus a lot of our other courses.”
Leaders themselves, says Arbogast, “will be disrupted or displaced” if they don’t upskill around AI but keeping soft skills sharp is all-important, too. “AI is learning and creating from the data being fed into it, finding the patterns and the networks, but EQ and the human skills of being able to interpret with nuance is true intelligence.”
“In a way, we have to redefine what creativity is,” says Williams. “A lot of people would say creativity separates humans from AI or machines. Yet generative AI on its own is more creative than most people. In the hands of a creative human, it’s even more so.”
