The 2022 State of AI in the Enterprise
Image credit: Depositphotos enhanced by CogWorld
Source: Irving Wladawsky-Berger, CogWorld Think Tank member
“Rapidly transforming, but not fully transformed - this is our overarching conclusion on the market, based on the fourth edition of our State of AI in the Enterprise global survey,” said Becoming an AI-fueled organization, the fourth survey conducted by Deloitte since 2017 to assess the adoption of AI across enterprises. “Very few organizations can claim to be completely AI-fueled, but a significant and growing percentage are starting to display the behaviors that can get them there.”
AI is increasingly viewed by workers as a trusted assistant. “Within just the last 18 months, AI capabilities have advanced considerably, maturing from what was often experienced as a bothersome critic - telling workers what to do or pointing out their mistakes - to more frequently serving as a copilot, independently executing on insights and trends surfaced through the power and speed of cloud-based data hosting and computation.”
These conclusions are similar to those of Stanford’s 2022 AI Index report, which found that AI was becoming more affordable and higher performing, with lower training costs and faster training times across a number of AI tasks including recommendation engines, image classification, object detection, and language processing. This has led to the widespread commercial adoption and increased real-world impact of AI systems.
Deloitte’s 2022 survey reached out to 2,875 executives from 11 advanced economies, to learn about the use of AI in their organizations, from their overall AI strategies and investments to their deployment of AI applications. Survey responses were classified based on the number of different AI applications that companies had deployed at scale, and the number of high achieving outcomes. The analysis revealed four key AI profiles:
Transformers - high deployments, high outcomes, - 28% of survey respondents. Transformers are the market leaders, with an average of 5.9 deployments of different AI applications at scale, and an average of 6.8 high achieving outcomes.
Pathseekers - low deployments, high outcomes, - 26% of respondents. Pathseekers only have 1.9 different AI application deployments at scale, but are achieving high outcomes, with an average of 6.2 applications.
Underachievers - high deployments, low outcome, - 17% of respondents. Underachievers have a significant number of application deployments underway, an average of 5.5, but haven’t yet adopted leading practices, and thus have an average of only 1.4 high achieving outcomes.
Starters - low deployments, low outcomes, - 29% of survey respondents. Starters are behind in building their AI capabilities, averaging only 1.6 full-scale deployments of different AI applications, and an average of 1.0 high achieving outcomes.
To ascertain the behaviors most associated with strong outcomes, Deloitte analyzed the survey data from each of these groups and conducted a number of executive interviews. The analysis and interviews revealed that these behaviors fell into four major categories: strategy, operations, culture and change management, and ecosystems. Let me summarize the key findings in each category.
Strategy - “AI-fueled organizations view AI as a key element of business differentiation and success, and they set an enterprisewide strategy that is championed from the top.”
38% of survey respondents, - ranging from 55% of transformers to 22% of starters - agree that their use of AI is key to differentiating their company from competitors, and 66% believe that their AI initiatives are critical to their future success.
However, only 40% of respondents said that their company has a well-defined enterprise wide AI strategy, with the percentage varying from 60% for transformers, 48% for pathseekers, 33% for underachievers and 19% for starters. Similarly, 40% of respondents said that their senior leaders communicate a bold vision for AI that will significantly change how their company operates, ranging from 57% for the highest achieving organizations, to 24% for the lowest.
“The strongest AI strategies tend to begin without ever mentioning AI.” Instead, they lean heavily on the organization’s core business strategy, coordinating and driving the AI strategy across the whole enterprise, working closely with all the business divisions in search of opportunities to leverage AI for competitive advantage. A successful AI strategy should aim to balance increased productivity with the creation of innovative new products and business models.
Operations - “AI-fueled organizations establish new operating models and processes that drive sustained quality, innovation, and value creation.”
“Technology cannot deliver transformative results unless organizations reimagine how work gets done.” The survey found that the most successful organization have reimagined their overall business workflows. 38% of respondents said that their functional group has undergone significant changes in business workflows to take advantage of new technology opportunities, including the creation of new AI jobs. As expected, the percentages ranged from around 55% for the highest performing groups, to 20%-25% for the lowest performing.
To be effective, these new operating models need to be carefully designed, documented, and followed across the organization, particularly the widely used machine learning operations processes (MLOps). While developing these AI processes and models is generally the responsibility of data scientists and IT experts, senior leaders should make sure that they are in place and adhered to across the business. On average, only about one third of survey respondents said that their functional group followed a documented AI model life cycle strategy and MLOps procedures when developing an AI solution, with the percentages ranging from about 50% for the highest performing group to around 10% for the lowest.
Culture and change management - “AI-fueled organizations nurture a trusting, agile, data-fluent culture and invest in change management to support new ways of working.”
“Over the past few decades, the pace of business and technology change has quickened, requiring workers to adapt, perpetually learn new skills, and make decisions amid growing ambiguity. For many organizations, these shifts have challenged a critical facet within their organization: their culture.”
Survey data and executive interviews showed that investments in change management have been key to a successful AI transformation, including data fluency and agility, while boosting trust and engagement across the organization. 37% of survey respondents said that their functional group invests in change management, incentives, or training activities to help people integrate new technology into their work, ranging from over 50% for the highest performing group, to around 25% for the lowest.
Since data is intrinsically intertwined with AI, raising the level of data literacy across the organization is key to AI success. While for some this involves advanced data capabilities, for the majority it means building the critical thinking skills needed to ask the right questions and the data fluency needed to find the right data to solve problems in their everyday work. Only around 25% of survey respondents said that they trust AI-derived insights more than their own intuition, with the highest performing group, the transformers, four times higher at 40% than the lowest group, the starters, at 10%.
Ecosystems - “AI-fueled organizations orchestrate dynamic ecosystems that help build and protect competitive differentiation.”
“No company has all the needed talent, algorithms, data sets, or breadth of perspective in-house to innovate perpetually with AI. That’s largely why most of today’s AI-fueled organizations establish robust technology ecosystems.” Around 75% of survey respondents said that they have two or more ecosystem relationships, with the highest-achieving groups, the transformers, at over 80% compared to around 60% for the lowest achieving groups.
“A healthier ecosystem approach typically identifies a base platform and looks for a variety of opportunities to integrate different vendors, including those that may be emerging or niche. When this approach is executed well, it not only protects from overdependence, but can also result in a higher level of differentiation, flexibility, and access to expanded perspectives on the market. Survey data reinforced this, showing that organizations with more diverse ecosystems were much more likely to have transformative visions for AI and use AI as a strategic differentiator.”
“Organizational progress has always depended on humans’ ability to imagine a new vision for the future and identify the opportunity available within it,” said the Deloitte survey in conclusion. “We seem to be rapidly approaching the day when AI could independently and reliably illuminate creative and strategic opportunity, releasing us from the confines of our limited perspectives. As we advance further into that AI-fueled future, those organizations that lay the foundations now will likely be rewarded manyfold.”
Irving Wladawsky-Berger is a Research Affiliate at MIT's Sloan School of Management and at Cybersecurity at MIT Sloan (CAMS) and Fellow of the Initiative on the Digital Economy, of MIT Connection Science, and of the Stanford Digital Economy Lab.