资源描述
kpmg 8 key AI adoption trends AI transforming the enterpriseIn this environment, staying competitive begins with a clear view of the AI landscape: where do some of the worlds largest and most influential companies stand in terms of AI investment, deployment, and real-world outcomes? We conducted the KPMG 2019 Enterprise AI Adoption Study to gain insight into the state of AI deployment efforts at select large cap companies. This involved in-depth interviews with senior leaders at 30 of the worlds largest companies, as well as secondary research on job postings and media coverage. These 30 highly influential, Global 500 companies represent significant economic value collectively, they employ approximately 6.2 million people, with aggregate revenues of $3 trillion. Together, they also represent a significant component of the AI market. We believe this group is an indicator of where very large enterprises are headed in terms of transforming their organizations with AI. To summarize the key learnings from our interviews and our insights from working with clients, we highlight eight key trends that can serve as guideposts for your organization. If your organization is lagging in AI deployment, these insights could help you move up the activation learning curve faster. If your organization is a leader in AI deployment, these insights may help fill in gaps and afford you greater confidence along your AI journey. Introduction U nderstanding the state of AI deploymenthow broadly it is being used and in what waysis challenging for many business leaders. Machine learning and other technologies are advancing significantly faster than many anticipated just a few years ago. The pace of development is accelerating and can be hard to grasp. 2019 KPMG LLP , a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative (“KPMG International”), a Swiss entity. All rights reserved.5 6 Introduction Internal governance emerging as key area The need to control AI Rise of AI-as-a-Service AI could shift the competitive landscape The trends Rapid shift from experimental to applied technology Automation, AI, analytics, and low-code platforms are converging Enterprise demand is growing New organizational capabilities are critical 1 2 3 4 8 7 AI transforming the enterprise 1 2019 KPMG LLP , a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative (“KPMG International”), a Swiss entity. All rights reserved.Whats at stake? Nearly all the executives we interviewed see AI as playing a role in creating new winners and losers. AI has broad enterprise applications and the potential to shift the competitive position of a business. The technologies under the AI umbrella are already contributing to product and service improvementsand they will be important drivers of innovation for wholly new products, services and business models. Focusing too narrowly on specific uses cases or cost savings can keep you from seeing and acting on the bigger picture: AI can be a game-changer that fundamentally shifts the competitive landscape. The winners will be the companies that use AI for the full range of benefits, from back and middle office productivity to front office product innovations and customer engagement. Against this backdrop, it is important to ask how AI can transform your business, and deploy the right technology, organizational capital, and data strategy to achieve that vision. Our research revealed valuable insights. For example, the five companies in our sample with more mature AI capabilities have an average of 375 full-time employees working on AI today, which includes data scientists, engineers, analysts and others. 1We estimate that, on average, each is spending $75 million on AI talent today. 2And they expect AI staffing numbers to continue to grow each company anticipates between 500 and 600 full-time employees working on AI in the next three years. Introduction 2 2 AI transforming the enterprise 2019 KPMG LLP , a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative (“KPMG International”), a Swiss entity. All rights reserved.Artificial Intelligence (AI): For the purpose of this research, we used a broad definition of AI. AI is a set of technologies that can imitate intelligent human behavior. Enterprise AI: Advanced intelligent digital systems at scale and as-a-service offered to business functions with oversight around the strategic direction of AI in the enterprise. This includes AI, machine learning, natural language processing (NLP), and other systems that conduct tasks that previously required human capabilities. Robotic Process Automation (RPA): An application of technology that enables organizations to configure computer software or a “bot” to interact with existing applications for processing a transaction, manipulating data, and communicating with other digital systems replicating manual work that typically would have been performed by a person. Intelligence is programmed or rule-based; not learned. RPA is often used in conjunction with AI to operationalize solutions within the current technology environment. Machine Learning (ML): An application of AI that provides systems with the ability to learn and improve from experience and training. Machine learning focuses on programs that can access data and use it to learn. Low-code platforms: Next generation technology platforms that leverage visual capabilities to build and deploy enterprise applications with embedded business rules, workflow, forms, and integration. Often in conjunction with RPA, these platforms also play a role in automation and enable enterprises to operationalize AI. Key definitions Research and methodology During the first half of 2019, KPMG conducted a two-part research study to better understand the state of AI deployment at large cap companies. Part one of the study consisted of secondary research analyzing job postings and media coverage for 200 of the Global 500 companies. In addition, researchers conducted interviews with three major technology companies that provide AI solutions and services (the supply-side). Insights gleaned from the secondary research informed the types of questions asked in part two of the research program. For part two, KPMG conducted in-depth interviews with senior executives at 30 of these Global 500 companies that buy and deploy AI (the demand side). Because organizational structure for AI varies by company, interviews involved a range of leaders responsible for AI deployment individuals interviewed include Chief Information Officers (CIOs), Chief Financial Officers (CFOs), Chief Data Officers (CDOs), and Chief Digital Officers (CDOs), as well as other senior line of business executives. Researchers used a semi-structured interview format, conducting 45-minute to one-hour interviews with each respondent. Following the interviews, researchers scored each company on a scale of 1 to 5, where 1 = low and 5 = very strong on a series of six criteria: governance, investment, pace, linkages, expertise, and management. A total of 30 points were possible. Companies were then coded into four ranked categories: Mature (24 to 30 points), Evolving (17 to 23 points), Aspirational (9 to 16 points), and Early (1 to 8 points). Lastly, interviews and secondary research were aggregated, then thematically coded. Overall, eight key trends clearly emerged from the data. The following is a deep dive into each of the trends. AI transforming the enterprise 3 2019 KPMG LLP , a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative (“KPMG International”), a Swiss entity. All rights reserved.The trends Less than three years ago, many leaders in large organizations were asking themselves if AI could support their productivity and growth objectivesand many were just beginning to pilot and test AI applications. But since then, we have seen rapid change. AI has quickly moved from a technology to watch to a technology to deploy. It is increasingly able to address a wide range of business issues, and companies across industries have successfully deployed automation at the functional level and within lines of business. This shift was driven by many factors including rapid digitization, advances in machine learning (ML) and the growing availability of data, which has enabled robust natural language processing (NLP) capabilities. Now, many enterprises are looking to convert pilot projects and drive benefits across their organizations. This involves driving scope (horizontal applications) as well as scale (vertical applications). Although the ability to blend both scope and scale is rare today, many of the large companies we interviewed are setting this as a core objective over the next three years: 3 Robotic Process Automation (RPA) Today, 26% of the companies we interviewed have deployed RPA at scale across the enterprise or major functions. Over half (65%) say their use of RPA today is selective and siloed by individual groups or functions. In three years, 83% expect to have RPA deployed at scale. AI and Machine Learning Today, only 17% of the companies we interviewed reported use of AI/machine learning at scale and 30% reported selective use in functions. In three years, half of the companies we interviewed expect to be using AI/machine learning at scale. Key takeaway The AI landscape has rapidly evolved in the last few years, and the technology is ready to deliver true value today. In the next three years, many enterprises are aiming to advance beyond initial, functional deployments towards true scale and scope through governance and center of excellence constructs. Rapid shift from experimental to applied technology 1 AI enables insight, augmentation and full automation Insights Capable of recognizing complex patterns from disparate sources of data and forming probabilistic insights Augmentation Software that can work alongside humans to learn patterns and augment human expertise Automation When combined with physical robots or software “bots,” full automation of complex tasks that typically involve human judgment is possible 4 AI transforming the enterprise 2019 KPMG LLP , a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative (“KPMG International”), a Swiss entity. All rights reserved.The trends Companies are deploying automation, AI, analytics, and low-code platforms in tandem, finding that these technologies work more effectively together. Key factors allowing companies to deploy these technologies together and take advantage of the benefits include: Rise of low-code platforms like Appian, OutSystems, Pega and ServiceNow that allow for integration and use of multiple technologies, and democratize the ability to code process automation software In-house talent that can work across many different technologies Use of best of breed integration tools like Mulesoft Leadership on point to ensure a coordinated approach across technologies Deploying automation, AI, analytics, and low-code capabilities in an integrated manner can support virtuous feedback loops: each supports the other, and together they go hand-in-hand with the scale and industrialization of AI. 4For example, think of the potential for super-charged analyticsmachine learning can enhance traditional analytics, making the process more data-intensive and enabling models that improve over time. This can lead to better predictions and outcomes. In our interviews, executives consistently pointed to the power of deploying automation, AI, analytics, and low-code capabilities together. In some organizations, this approach is driven from the top. But it can also emerge from existing teams. For example, many large companies have well-established global business service (GBS) teams. We heard that many of these GBS teams on their own see the value of bringing AI to their existing automation activities, effectively creating a GBS 2.0. GBS teams are often well-positioned to harness these technologies and drive new additional value since they already have extensive capabilities around enterprise services, deep understanding of the piping of different functional areas and extensive data repositories. The growing availability of AI and analytics is enhancing RPA and automation solutions and allowing some organizations to move from task automation to process automation, creating an entirely new level of business value. Key takeaway Look at automation, AI, analytics and low-code platforms as complementary technologies and services that can be mixed and matched to exponentially improve progress towards specific business goals. Automation, AI, analytics, and low-code platforms are converging Low-code platforms: Next generation technology platforms that leverage visual capabilities to build and deploy enterprise applications with embedded business rules, workflow, forms and integration. Often in conjunction with RPA, these platforms also play a role in automation and enable enterprises to operationalize AI. 2 AI transforming the enterprise 5 2019 KPMG LLP , a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative (“KPMG International”), a Swiss entity. All rights reserved. 2019 KPMG LLP , a Delaware limited liability partnership and the U.S. member firm of the KPMG network of independent member firms affiliated with KPMG International Cooperative (“KPMG International”), a Swiss entity. All rights reserved.The trends What is organizational capital? 3 Enterprise demand is growing In our interviews, we heard a consistent message: many large companies are making major investments in AI, aiming to vault their deployment over the functional-level bar and into other areas of their businesses. Investment in talent is a key requirement to move the AI agenda forward: The five companies in our sample with more mature AI capabilities have an average of 375 full-time employees working on AI today, which includes data scientists, engineers, analysts, and others. 5We estimate that, on average, each is spending $75 million on AI talent today. 6And they expect AI staffing numbers to continue to grow each company anticipates be
展开阅读全文