McKinsey report: Two AI trends top 2022 outlook

McKinsey’s newly-released Technology Trends Outlook 2022 named applied AI and industrializing machine learning as two of 14 of the most significant technology trends unfolding today. 

According to McKinsey, the study builds on trend research shared in 2021, adding new data and deeper analysis and examining “such tangible, quantitative factors as investment, research activity, and news coverage to gauge the momentum of each trend.”

Applied AI tops list with maturity and innovation

Applied AI, considered by McKinsey as based on proven and mature technologies, scored highest of all 14 trends on quantitative measures of innovation, interest and investment, with viable applications in more industries and closer to a state of mainstream adoption than other trends. 

In a 2021 McKinsey Global Survey on the state of AI, 56% of respondents said their organizations had adopted AI, up from 50% in the 2020 survey. According to the 2022 report, tech industries are leading in AI adoption, while product development and service operations are the business functions that have seen the most benefits from applied AI.

Roger Roberts, partner at McKinsey and one of the report’s coauthors, said of applied AI, which is defined “quite broadly” in the report, “We see things moving from advanced analytics towards… putting machine learning to work on large-scale datasets in service of solving a persistent problem in a novel way,” he said. 

That move is reflected in an explosion of publication around AI, not just because AI scientists are publishing more, but because people in a range of domains are using AI in their research and pushing the application of AI forward, he explained. 

“There is really that path from science, to engineering, to scale,” he said. “We see AI moving quite quickly down that path, and what I’m really excited about is the fact that more things are moving from engineering to scale.” 

However, the McKinsey report also highlighted a variety of key uncertainties that could affect the future of applied AI, including the availability of talent and funding, cybersecurity concerns and questions from stakeholders about the responsible and trustworthy use of AI.  

McKinsey says industrializing AI is a growing trend

According to the McKinsey report, industrializing machine learning (ML) “involves creating an interoperable stack of technical tools for automating ML and scaling up its use so that organizations can realize its full potential.” The report noted that McKinsey expects industrializing ML to spread as more companies seek to use AI for a growing number of applications. 

“It does encompass MLops, but it extends more fully to include the way to think of the technology stack that supports scaling, which can get down to innovations at the microprocessor level,” said Roberts. “You’re seeing lots of new capabilities in silicon that support the acceleration of particular classes of AI work, and those innovations will move into broader use, allowing for faster and more efficient scaling both in terms of computing resources, but also more sustainability.” 

The report cites software solutions corresponding to the ML workflow, including data management, model development, model deployment and live model operations. It also includes integrated hardware and heterogeneous computing used in ML workflow operations. 

Roberts added that he sees big tech organizations such as Google, Meta and Microsoft as in the lead on industrialized ML “by a longshot.” But he predicted the trend would soon make its way well beyond those companies: “We’ll start to see more and more venture activity and corporate investment as we build that tool chain for this new class of software and this new class of product as productized services,” he explained. 

McKinsey predicts continued AI momentum

Roberts emphasized that in his view, economic issues won’t change AI’s powerful momentum.  

“There’s never been a better time to be leading the application of AI to exciting business problems,” he said. “I think there’s enough momentum and capability flowing along the path of science to engineering to scale.” He did add, however, that within industries there may be some growing separation of leaders and laggards. 

“Leaders will continue to make the right investments in talent tooling and capabilities to help deliver scale,” he said. “Laggards may let the opportunity slip away if they’re not careful.”