The use of artificial intelligence (AI) technologies in businesses is set to grow, but CIOs should focus on using pilot projects to ready their businesses to use the technology, information technology market research multinational Gartner says.
Its ‘2018 CIO Agenda Survey’ found that 46% of CIOs plan to deploy AI systems, while only 4% of CIOs have AI systems in place.
“Early adopters are facing many obstacles to the progress of implementing AI in their organisations,” says Gartner research VP Whit Andrews.
Gartner predicts that, up to 2022, 85% of AI projects will deliver erroneous outcomes because of bias in data, algorithms or the teams responsible for managing them.
“Expect AI projects to produce, at best, lessons that will help with subsequent, larger experiments, pilots and implementations,” Andrews adds.
Most organisations are not well prepared for implementing AI because they lack internal skills in data science and plan to rely on external providers to fill the gap.
“Relying on external suppliers for data science skills is not an ideal long-term solution. Data is the fuel for AI. Organisations need to prepare to store and manage large amounts of data for AI initiatives,” says Gartner research VP Jim Hare.
However, of the CIOs surveyed, 53% rated their own ability to mine and exploit data as limited, which is the lowest level.
Therefore, Hare suggested that companies ensure that early AI projects transfer knowledge from external experts to their employees, as well as build up the organisations’ in-house capabilities before moving on to large-scale projects.
“While there is potential for strong growth as CIOs begin piloting AI programmes through a combination of buy, build and outsource efforts, companies should not aim to achieve hard outcomes, such as direct financial gains, with AI projects,” says Andrews.
It is best to start AI projects with a small scope, such as process improvements, customer satisfaction or financial benchmarking. Understand what you are trying to accomplish on a small scale, and only then pursue more dramatic benefits, he advises.
“Whether an AI system produces the right answer is not the only concern. Executives need to understand why it is effective and external service providers must offer insights when it is not.”
It is important that some insight into how decisions are reached is built into any service agreement, emphasises Andrews.
“Although it may not always be possible to explain all the details of an advanced analytical model, such as a deep neural network, visualisation of the potential choices is necessary. In situations where decisions are subject to regulation and auditing, it may be a legal requirement to provide this kind of transparency,” he adds.