英国人工智能创新方向报告：基于新数据库的战略洞见及改革路径

THE DIRECTIONOF AI INNOVATIONIN THE UK INSIGHTS FROM A NEW DATABASEAND A ROADMAP FOR REFORM Carsten JungandBhargav SrinivasaDesikanApril 2025 ABOUT IPPR IPPR, the Institute for Public Policy Research, is an independent charityworking towards a fairer, greener, and more prosperous society. We areresearchers, communicators, and policy experts creating tangible progressivechange, and turning bold ideas into common sense realities. Working acrossthe UK, IPPR, IPPR North, and IPPR Scotland are deeply connected to thepeople of our nations and regions, and the issues our communities face. We have helped shape national conversations and progressive policy changefor more than 30 years. From making the early case for the minimum wageand tackling regional inequality, to proposing a windfall tax on energycompanies, IPPR’s research and policy work has put forward practicalsolutions for the crises facing society. IPPR4th floor,8 Storey's GateLondonSW1P 3AY E: info@ippr.orgwww.ippr.orgRegistered charity no: 800065 (England and Wales),SC046557 (Scotland)This paper was first published in April 2025. © IPPR 2025The contents and opinions expressed in this paper are thoseof the authors only. The progressive policy think tank CONTENTS Summary..........................................................................................................................51. Introduction: AI deployment needs not just acceleration, but direction....8Measuring the direction of innovation and identifying‘deployment gaps’.................................................................................................92. Key findings from our new database.................................................................10AI businesses focus mainly on general process improvementsrather than specific problem solving.............................................................10AI adoption is focussed on the knowledge economy andprocess improvement........................................................................................ 11The use of off-the-shelf models: AI adoption in the UK couldto a large extent involve business process innovation ratherthan AI software innovation per se................................................................143. Deep dives: What are the value propositions of AI companiesand where are the gaps?...................................................................................... 17Case study 1: significantly improving public health will requiremore focus on prevention................................................................................. 17Case study 2: Transport AI innovation has a big focus onautonomous vehicles and logistics but not on improving access..........204. Policy recommendations for mission-driven AI innovation.........................23Recommendation 1: The government needs to better trackAI deployment to inform policy.......................................................................24Recommendation 2: Break missions down to specificproblem statements..........................................................................................25Recommendation 3: Align innovation policy clearly withmissions to create ‘technology push’...........................................................26Recommendation 4: Use subsidies, procurement andpreferential financing for AI adoption and marketshaping, creating ‘demand pull’......................................................................30References.....................................................................................................................34Appendix: Methodology.............................................................................................37 ABOUT THE AUTHORS Carsten Jungis head of AI at IPPR. Bhargav Srinivasa Desikanwas a senior research fellow at IPPR at the time of writing. ACKNOWLEDGEMENTS We would like to thank Annie Williamson and Stephen Frost for their in-depthadvice on the health and transport sector case study. We would also like to thankAlexandra Lowe, Stuart Thompson, Kieran Neild-Ali, Charles McIvor, Sam Freedman,Kir Nuthi, Peter Hyman, Harry-Quilter Pinner, George Dibb, Simone Gasperin, RainerKattel, Seb Krier and Andrew Bennett for very helpful conversations and comments.We would also like to thank Sylvia Monkhouse, Richard Maclean and Abi Hynes andfor copyediting and producing the report to the highest standard. Download This document is available to download as a free PDF and in other formats at:http://www.ippr.org/articles/the-direction-of-ai-innovation-in-the-uk Citation If you are using this document in your own writing, our preferred citation is:Jung C and Srinivasa Desikan B (2025)The direction of AI innovation in the UK: Insights from a new databaseand a roadmap for reform, IPPR. http://www.ippr.org/articles/the-direction-of-ai-innovation-in-the-uk Permission to shareThis document is published under a creative commons licence: Attribution-NonComme