What’s the current state of AI? I’ve been closely following AI over the past several years, and based on everything I’ve read and written about in my blog, the question brings to mind Charles Dickens’ A Tale of Two Cities.
We have in fact a tale of two very different AI worlds. Ever since OpenAI released ChatGPT a few months ago and encouraged the general public to try it, we’ve seen article after article about its emergent unexpected capabilities and potential existential risks. But, at the same time, most recent surveys of the state of AI in the enterprise have concluded that the majority of surveyed companies are still in the early stages of deployment and have not achieved significant outcomes.
For example, Stanford’s recently published 2023 AI Index Report found that industry was now leading the development of state-of-the-art large language models like GPT-4 and chatbots like ChatGPT. But at the same time, the Stanford report references the most recent McKinsey global survey on AI which found that while leading edge firms continue to pull ahead, the number of companies adopting AI has plateaued.
The global survey, “The State of AI in 2022 — and a Half Decade in Review,” marks the fifth consecutive year in which McKinsey has analyzed AI’s impact on companies around the world. The 2022 report is based on a survey of almost 1,500 firms from a variety of regions, industries, and company sizes. Only about half of respondents said that their organization have adopted AI in at least one function.
Let me summarize the report’s key findings.
AI adoption has more than doubled over the past five years. In 2017, 20% of respondents said that they had adopted AI in at least one business area. At 50%, the proportion of companies that have adopted AI in 2022 has more than doubled, but it has plateaued in recent years at between 50% and 60%.
The average number of AI capabilities adopted by organizations has also doubled from 1.9 in 2018 to 3.8 in 2022. The most commonly deployed AI capabilities are robotics process automation, — mentioned by 39% of respondents; computer vision — 34%; natural-language text understanding —33%; conversational interfaces — 33%; and deep learning — 30%.
The top use cases have remained fairly stable over the past few years, with service operations optimization at 24%; new AI-based products — 20%; customer service analytics — 19%; customer segmentation — 19%; AI-based enhancement of products — 19%; and customer acquisition — 17%.
AI investment has increased alongside its rising adoption. In 2018, 40% of respondents using AI reported that more than 5% of their digital budgets went to AI. In 2022, the equivalent figure was 52%. And 63% of respondents expect that their organization’s investments in AI will increase over the next three years.
Organizations that have adopted AI realized meaningful revenue and costs benefits. In 2021, 63% of organizations that adopted AI reported a significant revenue impact in a number of areas, with marketing and sales mentioned by 70%; product and service development — 70%; strategy and corporate finance — 65%; supply chain management — 63%; and manufacturing — 61%.
At the same time, 32% of organizations reported significant cost benefits, with supply chain management — 52%; service operations — 45%; risk management — 43%; strategy and corporate finance — 43%; and manufacturing — 42%.
While AI use has increased, there have been no substantial increases in the reported mitigations of AI-related risks. The risks that AI-adopting organizations consider most important to mitigate are cybersecurity — 51%; regulatory compliance — 36%; personal & individual privacy — 28%; explainability — 22%; and organizational reputation — 22%.
The 2022 McKinsey survey also found that AI high performers have expanded their competitive advantage over the past five years and closely examined what these AI leaders do differently.
The proportion of AI high performers has remained steady at about 8%. McKinsey defined AI high performers as those organizations that are seeing 20% or more bottom-line impact from their adoption of AI as measured by Earnings Before Interest and Taxes (EBIT). These AI leaders are achieving their superior results mainly through increased top-line revenues rather than through cost reductions, although AI had also helped decrease their costs.
High performers are more likely to leverage AI in their key business practices. Their top business practices include linking AI initiatives to business value across the organization; having an AI strategy that’s aligned with the overall corporate strategy; a clearly defined AI vision and strategy; and a senior management team that’s committed to the organization’s AI strategy.
In addition, AI high performers led in the development and deployment of AI at scale across the organization. “In the past year, high performers have become even more likely than other organizations to follow certain advanced scaling practices, such as using standardized tool sets to create production-ready data pipelines and using an end-to-end platform for AI-related data science, data engineering, and application development that they’ve developed in-house.”
High performers also lead in the management of AI-related risks like personal privacy, and equity and fairness. They’re also more likely to engage in risk mitigation practices like data governance, standardized processes and protocols, and automated data quality control.
AI high performers continue to outspend other organizations. High performers are nearly eight times more likely than others to say that they spend at least 20% of their digital-technology budgets on AI-related projects. In addition, they’re over five times more likely to report that their organizations spend more than 20% of their enterprise-wide revenue on digital technologies.
The McKinsey report also includes a detailed look at the overall AI talent picture, including the strategies organization use for acquiring the necessary talent.
Organizations have largely shifted from experimenting with AI to actively embedding it in enterprise applications. Survey responses show that the AI skills organizations are hiring most often are software engineers — 39%; followed by data engineers — 35%; AI data scientists — 33%; machine learning engineers — 30%; and data architects — 28%.
Hiring is a challenge, but less so for high performers. “All organizations report that hiring AI talent, particularly data scientists, remains difficult. AI high performers report slightly less difficulty and hired some roles, like machine learning engineers, more often than other organizations.” High performers are particularly focused on hiring for AI deployment and business value optimization.
The tech talent shortage shows no sign of easing, with a majority of respondents reporting difficulty in hiring AI-related skills. AI data scientists remain particularly scarce, with 32% of survey respondents saying that hiring them was very difficult; followed by machine learning engineers, — 28%; machine translation specialists — 27%; and AI data architects — 25%.
AI high performers report sourcing AI-related talent in a broader variety of ways than other organizations. The top source of AI talent for high performers are top-tier technical universities — 58%; followed by reskilling of internal employees — 47%; top-tier global technology companies — 46%; other technology companies — 39%; and other universities — 37%.
Finally, the 2022 McKinsey report explored the level of diversity in AI and concluded that there’s significant room for improvement at most organizations. The average share of women in AI teams is 27%, and the average share of racial or ethnic minorities in AI teams is 25%. 46% of AI respondents said that their organizations have active programs to increase gender diversity, while 33% said that they have similar programs to increase racial and ethnic diversity.
As in previous McKinsey studies, the research shows that there’s a correlation between diversity and AI performance. Organizations that said that at least 25% of employees involved in AI are women are 3.2 times more likely than others to be AI high performers, and organizations at which at least 25% are racial or ethnic minorities are more than twice as likely to be AI high performers.
“Over the past half decade, during which we’ve been conducting our global survey, we have seen the AI winter turn into an AI spring,” said McKinsey Global Institute Parter Michael Chui. “However, after a period of initial exuberance, we appear to have reached a plateau, a course we’ve observed with other technologies in their early years of adoption. We might be seeing the reality sinking in at some organizations of the level of organizational change it takes to successfully embed this technology.
Comments