AI, complicated but beneficial
Many business leaders are excited about artificial intelligence, but when it comes to investing in the technology, they may be entranced, bothered, and confused. It doesn’t have to be like this. We surveyed people who work closely with AI solutions and asked them to share their experiences on the best approaches to sell AI 00 the right AI to business leaders at the right time.
First of all, it’s imperative to show “cognitive politeness,” advises Bjorn Austraat, senior vice president and head of AI acceleration at Truist. “Invest the time and effort upfront to translate difficult technical concepts into plain English. Likewise, create a compelling, properly articulated vision for all project stakeholders to ensure alignment and productive collaboration.”
Austraat explains that he uses a “simple phrase” to push the vision of AI. Ask, “’Whose life will improve, by how much, and how do we know?’” he says. “If you can really answer that question, then you’ve got all the bases covered, from design to deployment and value proposition to political air cover.”
This clear vision is strongly recommended by other industry leaders. “The key to selling AI is having a clear, simple explanation of the business case,” agrees Ben Hookway, CEO of Relative Insight. “With consumer data pouring in from all directions, much of it is text-based. All of this qualitative text data can – and should – have a strategic impact on brand strategy, product development, customer care, marketing, communications and more. Much of this data is real-time and extremely valuable in driving a true understanding of how customers are feeling.”
In Australia’s financial services industry, “explainability trumps model performance,” he adds. “In particularly sensitive areas such as lending, banks and other financial services institutions must balance the desire for innovation and the use of cutting-edge AI with reasonable regulatory expectations of explainability, robustness and fairness. The hottest solution doesn’t always win, especially when it’s too much of a black box.”
It’s also important to convince executives that successful AI can take the business in new — and potentially positive — directions. “Enterprises don’t want to buy sensors, cameras or AI solutions – they’re looking for a total solution that solves a business problem,” said Ken Mills, CEO of EPIC iO. “If you’re trying to sell sensors or video cameras to businesses, my advice is to partner with consultants or solution providers who can provide comprehensive answers to business problems.”
AI is about simplifying processes,” says Umesh Sachdev, co-founder and CEO of Uniphore. “Whether it’s solving customer problems, eliminating manual tasks, or making more accurate predictions, AI complements human effort so people can work, interact, and innovate more efficiently.”
It goes even further – with the advent of far-reaching AI solutions like conversational AI that “give customers and employees access to advanced problem-solving capabilities, the world as we know it will be fundamentally changed,” writes Robb Wilson, founder of OneReach .ai in his latest book, Age of Invisible Machines. Business leaders need to understand that AI cannot thrive over the long term as separate, one-off projects. This “requires a holistic endeavor that touches all aspects of your business.” like deploying disparate machines that exist in isolation – will underwhelm your employees and customers, resulting in low adoption rates. However, a fully integrated approach can create a whole new paradigm of productivity with unprecedented potential.”
Ultimately, the potential is enormous, which can amaze even the most AI-savvy individuals. “Creative automation used to be a pretty magical innovation in digital marketing,” says Peter Gordon, global head of AI product at Hogarth Worldwide. “But nothing quite compares to typing the most bizarre sentence imaginable – something like ‘teddy bear on a skateboard in the middle of Times Square’ – and seeing AI instantly generate a perfect photorealistic image of your idea. Or speaking to the chatbot of an impressively perceptive Steve Jobs that really feels like a face-to-face conversation. And there is so much more to come.”
Overcoming executives’ nervousness about AI is also key to successfully selling the technology. People’s first reactions to AI “may be a kind of magical fear,” says Gordon. It’s important for business leaders to understand that AI technology “enables them to do what they do best — accelerate the ideation processes to create on-brand ideas never seen before.”
Where and how to start? Keep it clear and simple, advises Wilson. “Often the easiest way to get started is to automate internally first; Start small by automating individual tasks and skills, not entire jobs. The simpler you make your starting point, the easier it is for you to test and iterate. The sooner you test and iterate, the sooner you can implement an internal solution. They continue to test and iterate on this solution, using the momentum to find new skills to develop, test, iterate and deploy. You fidget a lot when you grow legs, but that’s part of the game.”
AI is successful when it addresses distinct business problems or creates new opportunities. “When companies are considering AI, it’s because they have a need,” says Sachdev. “A process is draining their resources or affecting their relationship with their customers, and they want to rectify this as easily and inexpensively as possible. Look for friction points. Which areas within a business process slow down or unduly burden customers or employees? What outcomes does a company want to improve and what is preventing them from achieving them?”
For example, friction points that can be addressed by AI are associated with the customer experience. “If customers are struggling on the frontend, perhaps AI can guide them by curating relevant content based on their past purchase and search histories,” says Sachdev. “When it comes to customer support, there is a wealth of avenues to explore. AI can help identify customer intent, improve self-service interactions, and even assist agents with live interactions. It can check conversations for compliance and analyze them for quality assurance. Either way, it saves human workers time and effort.”
