Artificial intelligence (AI) could help to address sluggish productivity growth in OECD countries. This book provides evidence for policymakers, business leaders, and researchers to help understand the adoption of AI in enterprises and the policies needed to enable this. The core analysis draws on a new policy-oriented survey of AI in enterprises across the Group of Seven (G7) countries and Brazil, complemented by interviews with business representatives. The book offers a comprehensive examination of barriers to the use of AI and examines actionable solutions, including in the areas of training and education, qualification frameworks, public-private research partnerships, and public data. Also examined is the work of public institutions that seek to facilitate the diffusion of digital technologies, including AI. Further, this book highlights the need for better policy evaluation, greater international comparability in surveys of AI, and studies of generative AI in business (widespread interest in which began after this survey).
The Adoption of Artificial Intelligence in Firms
Abstract
Executive Summary
This study combines several types of data and information to explore the adoption of artificial intelligence (AI) in enterprises and how governments can support this. The core of the study is a policy-oriented survey of 840 enterprises implemented in the Group of Seven (G7) countries – the 2022-23 OECD/BCG/INSEAD Survey of AI-Adopting Enterprises - plus 167 enterprises in Brazil. The survey includes novel questions on topics such as enterprises’ views on the value of public policies relevant to AI uptake, and their priorities for future AI policy. Other novel questions seek information on familiar topics, such as enterprises’ use of cloud computing, but with new emphases – such as probing why cloud computing, an important adjunct to AI, might be underutilised. Complementing the survey, this book also contains studies of public sector institutions that help technology diffusion, along with interviews with enterprises.
Achieving higher rates of AI adoption could raise labour productivity and have other desirable outcomes, such as lower defect rates in production, reducing the need for material inputs. Policy insights from this and other comparable studies will likely become more important as more enterprises seek to become active users of AI.
It is widely reported that a scarcity of skills – particularly specialised talent – hinders AI uptake, even in many large firms. This study shows that policies and programmes to develop human capital are among the most valued and used by businesses. Many enterprises express a desire to better understand how to identify and use the right AI skills. However, academic certifications may not provide all the information employers seek. Updated qualification frameworks could help. Public sector providers should collaborate with industry to design relevant training materials, and training programmes should be tailored to industry or business-specific needs (such as using AI to optimise supply chain management). Training on real-world projects, using AI systems and datasets common in specific areas of business, is particularly valuable.
Policy makers need to prioritise the quality of data in public repositories, for example by removing outdated or conflicting information. They should strive to simplify procedures for acquiring public data wherever possible. To aid users in understanding the data's meaning and context, comprehensive documentation should be made readily available. Furthermore, establishing centralised platforms for accessing public sector data could streamline search and retrieval processes. Policy makers should also seek to enhance the compatibility of legal frameworks governing cross-border data flows. International data sharing can pose severe challenges for companies that operate in multiple jurisdictions.
To help use and develop AI, enterprise collaborations with universities and public research institutions are widespread, highly valued and have several purposes. Some public financial support for collaborations, especially the first experience, could help to mitigate risks and reduce hesitancy among some firms. Research and development (R&D) tax incentives can also be designed to encourage industry-university collaboration and research commercialisation.
The process of applying for public funds that aid collaborative AI research should be straightforward. Information could be made widely available describing funding opportunities, evaluation criteria and examples of successful applications. Firms and universities could benefit from the development of model framework agreements for collaboration. Additionally, universities and public research organisations should ensure transparency in their key operational practices, for instance, in terms of overall project governance.
Most OECD countries have public institutions dedicated to facilitating firms’ uptake of digital technologies, including AI. These institutions frequently highlight uncertainty over the return on investment as a critical obstacle for firms considering adopting AI. They emphasise that a lack of data maturity is a fundamental barrier to implementing AI. Additionally, they report that managers often struggle to understand how AI can address real problems in the workplace and simultaneously underestimate the enterprise-wide implications and changes in business culture that AI may entail.
These institutions implement many types of initiative, from developing proofs-of-concept demonstrating how AI can help firms, to creating networking and collaborative platforms to help build AI ecosystems of public and private actors. A significant share of enterprises has used and positively values various public services to aid the adoption of AI. The diverse and sometimes innovative designs of these initiatives across institutions and countries offer opportunities for policy learning.
Even though many of the sampled enterprises use AI in advanced ways, they still seek additional information on several domains of AI. This suggests that policy makers should look for cost-effective ways of signposting and/or providing easily findable, accessible, current, and specific information and advice, for instance, on regulatory updates, compliance guidance and evolving business use cases for AI.
Governments could also provide guidelines or a framework to help small and medium-sized enterprises navigate the vendor selection process, indicating, for instance, important considerations to be aware of when choosing an AI vendor. Guidelines outlining agency roles and expertise, along with mechanisms for companies to communicate their needs, would also help. Enterprises also wish to see clearly stated accountability frameworks for the safe use of AI. Policy makers need to examine regulations for ambiguities and assess how best to communicate regulatory information to firms.
Better international comparability among surveys of AI in firms would help policy making. Several national statistical offices helped to shape the 2022-23 OECD/BCG/INSEAD survey questionnaire, and several of its new questions might be considered for inclusion in future national surveys.
Policy makers should examine the cost, scale and impact of diffusion institutions’ work as well as related policies towards the uptake of AI. Most such institutions only work with a limited number of client firms. Key questions to address include whether these bodies generate wider demonstration effects and, if so, the magnitude of such secondary impacts. Such analyses would help in shaping efforts to diffuse AI more widely.
