Example: bankruptcy

The macroeconomic impact of artificial intelligence

The macroeconomic impact of AI Technical report of artiazakhstan Draft for discussion Confidential October 2017 February 2018 The macroeconomic impact of artificial intelligence 1. Executive summary 1 2. What is AI and how could it impact the economy? 6 3. Overview of our approach 14 4. Econometric analysis 18 5. Job automation study 26 6. AI impact Index 31 7. S-CGE model analysis 35 8. Sensitivity analysis 56 9. Conclusion 62 10. Existing and forthcoming insights on AI insights from PwC 64 Contents Appendices 65 macroeconomic impact of artificial intelligence 1 Purpose of this report With artificial intelligence (AI) set to transform the way that we live and work, it raises the inevitable question of how much the technologies will impact businesses, consumers and the economy more generally.

Macroeconomic impact of artificial intelligence 2 1.2. Research methodology Estimating the global economic impact of AI is a pressing and challenging task. As a result, we have used a complex, multi-stage modelling approach that focussed on …

Tags:

  Macroeconomics

Information

Domain:

Source:

Link to this page:

Please notify us if you found a problem with this document:

Other abuse

Transcription of The macroeconomic impact of artificial intelligence

1 The macroeconomic impact of AI Technical report of artiazakhstan Draft for discussion Confidential October 2017 February 2018 The macroeconomic impact of artificial intelligence 1. Executive summary 1 2. What is AI and how could it impact the economy? 6 3. Overview of our approach 14 4. Econometric analysis 18 5. Job automation study 26 6. AI impact Index 31 7. S-CGE model analysis 35 8. Sensitivity analysis 56 9. Conclusion 62 10. Existing and forthcoming insights on AI insights from PwC 64 Contents Appendices 65 macroeconomic impact of artificial intelligence 1 Purpose of this report With artificial intelligence (AI) set to transform the way that we live and work, it raises the inevitable question of how much the technologies will impact businesses, consumers and the economy more generally.

2 Employees want to know what AI means for their job and income, while businesses are asking how they can capitalise on the opportunities that AI presents and where investment should be targeted. Cutting across all these considerations is how to build AI in the responsible and transparent way needed to maintain the confidence of customers and wider stakeholders. Traditionally, the research into the impact of AI, such as Frey and Osborne (2013) and Autor (2003), has focused on the effects on employment, as some jobs and tasks become automated and firms seek to make their business run more efficiently.

3 More recently, some authors have focused on the benefits that could come from productivity gains associated with this automation. However, the possible benefits and opportunities of AI go much further. The ability to collect, store and analyse data at the scale, speed and in the ways facilitated by AI technologies will allow firms to improve the quality of products and tailor them to consumers, increasing their value. AI can also reduce the amount of time that consumers spend doing low-value tasks or reduce frictions in the consumption process, all leading to increased consumer demand. We seek to provide a clearer picture of the full economic potential of AI globally, extending the exploration of AI s potential beyond the simple replacement of workers, to AI that augments the workforce and productivity.

4 We also explore the impact of AI-driven consumption-side product enhancements on the economy, which, to our knowledge, has not yet been explored in any great detail within the AI literature. These are the topics that we discuss and address in a series of PwC reports. In June 2017 we published our report, Sizing the prize: What s the real value of AI for your business and how can you capitalise?1, which highlighted how AI can enhance and augment what enterprises can do and provides a clear and compelling case for AI investment and development. In this report, we provide detailed insight into the approach that we took to complete the analysis of the global economic impact , as well as a more in-depth look at the results of our analysis and an exploration of the robustness of those results.

5 Our research has already created numerous insights into AI s possibilities and potential impacts in different sectors and regions. We are using these insights to help our clients leverage them in an effective, efficient and intelligent way helping distinguish them from their competition and ensure they are ready for the age of AI. Our study not only captures AI s impact through more channels than previously covered in a single study, but also presents detailed findings on both the geographic and sectoral distribution of these results. We look at which regions are set to gain more or less, and also examine in detail how the different AI-driven impacts on the economy will unfold over time in practice.

6 Whilst we place a lot of confidence in the robustness of our main scenario figures, we have also run a number of scenario-based sensitivity tests to test not only whether our least robust assumptions had a material impact on the results, but also to provide insight into a world where AI impacts the economy differently due to heterogeneous possible business and consumer responses to AI s introduction. The following subsections discuss the research methodology, key findings and sensitivity tests we conducted as part of this study. 1 This study formed part of a series of PwC publications related to artificial intelligence which address the key questions discussed in this section.

7 1. What AI means for jobs and income 2. How to build AI in a responsible and transparent way 3. How businesses can capitalise on the opportunities 1. Executive summary macroeconomic impact of artificial intelligence 2 Research methodology Estimating the global economic impact of AI is a pressing and challenging task. As a result, we have used a complex, multi-stage modelling approach that focussed on leveraging the global PwC network s resources and research capabilities to provide sufficiently accurate and convincing answers to such a compelling topic. Our approach follows three stages.

8 Figure summarises each of these stages and the types of analysis that underpin each. The first two stages involved undertaking primary research aimed at estimating the relationship between AI and both consumer products and firm productivity, before combining these with estimates of the pace and profile of AI adoption to evaluate the direct impacts of AI on each of these two elements. We used existing PwC research into the potential for AI-driven job automation as well as new econometric analysis assessing the relationship between AI and labour productivity to identify the key drivers of productivity growth, understanding where AI fits into this picture and specifying models capable of picking up the true causal effect of AI on productivity.

9 PwC s AI impact Index, developed by our AI experts in partnership with Fraunhofer, scored different product lines according to five key criteria especially developed to evaluate AI s impact on products. In particular, the research captured the potential that AI has to improve the quality of products, the potential for products in an industry to be more personalised and the amount of time that consumers could save from using AI. The final stage focussed on bringing the analysis together and converting these results into AI-driven inputs into our Spatial Computable General Equilibrium (S-CGE) model2 a dynamic model of the global economy that we have used to estimate the net global impact of AI on the economy up until 2030.

10 Beyond the initial impacts on productivity, job displacement and consumer choice, these net effects account for secondary impacts such as the creation of new jobs, increased consumer demand (from attractive goods first and more affordable goods second), the increased supply of labour to the market, and trade flow patterns. Figure : Our multi-stage approach to assessing the total economic impact of AI Analysing productivity impacts:Top-down econometrics analysisCapturing AI-enabled product enhancements: PwC s AI impact indexEstimating levels of automation and augmenting AI uptakePace and profile of adoptionAutomationprojectionsAugmenting AI projectionsEconomic impactEmploymentGDPCGE model analysis[ ][ ][ ][ ]12 Source.


Related search queries