Title PART 71.pdf ✅

58 Finance & economics The Economist February 15th 2025 ChatGPT did not make this chart 1 "In which specific tasks is Al most useful*?" United States, Aug 2024,% responding 0 10 20 30 40 50 60 Written communication Searching for information Documentation Interpreting/translating Administrative tasks Coding Data analysis Customer support Idea generation Tutoring Other *At work in the previous week Source: A. Bick, A. Blandin and D. Deming, CEPR nia and co-authors found a general-pur- pose AI tool improved the quality of legal work, such as drafting contracts, most no- tably for the least talented law students. The problem is that this is swamped by another effect. Ajob can be considered as a bundle of tasks, which tech may either commoditise or assist with. For air-traffic controllers, tech is an augmentation: it processes flight data while leaving deci- sions to humans, keeping wages high. By contrast, self-check-out systems simplify cashiers' roles, automating tasks such as calculating change. This lowers the skill requirement, causing wages to stagnate. Thus despite the early optimism, cus- tomer-service agents and other low-skilled workers may face a future akin to cashiers. Their repetitive tasks are susceptible to automation. Amit Zavery of ServiceNow, a business-software company, estimates that more than 85% of customer-service cases for some clients no longer require human involvement. As AI advances, this figure will probably rise, leaving fewer agents to handle only the most complex cases. Although AI may at first boost pro- Imagine Archimedes with Al 2 United States, impact of generative Al on new material discovery*, % change 100 75 50 25 95% confidence ductivity, its long-term impact will be to commoditise skills and automate tasks. Unlike earlier automation, which re- placed routine jobs such as assembly-line work and book-keeping, AI may extend its reach to non-routine and creative work. It can learn tacitly, recognise patterns and make predictions without explicit instruc- tion; perhaps, in time, it will be able to write entertaining scripts and design use- ful products. For the moment it seems as though, in high-wage industries, it is junior staff who are the most vulnerable to auto- mation. At A&O Shearman, a law firm, AI tools now handle much of the routine work once done by associates or paralegals. The company's software can analyse contracts, compare them with past deals and suggest revisions in under 30 seconds. Top per- formers have been best at using the tech to make strategic decisions, says David Wakeling, the firm's head of AI. The shift in recent economic research supports his observation. Although early studies suggested that lower performers could benefit simply by copying AI outputs, newer studies look at more complex tasks, such as scientific research, running a business and investing money. In these contexts, high performers benefit far more than their lower-performing peers. In some cases, less productive workers see no im- provement, or even lose ground. Intelligent design Aidan Toner-Rodgers of MIT, for instance, found that using an AI tool to assist with materials discovery nearly doubled the productivity of top researchers, while hav- ing no measurable impact on the bottom third. The software allowed researchers to specify desired features, then generate candidate materials predicted to possess these properties. Elite scientists, armed with plenty of subject expertise, could identify promising suggestions and dis- card poor ones. Less effective researchers, by contrast, struggled to filter useful out- puts from irrelevant ones (see chart 2). Similar results have emerged in other areas. Nicholas Otis of the University of California, Berkeley, and co-authors found that stronger Kenyan entrepreneurs raised their profits by over 15% with an AI assis- tant, and strugglers saw profits fall. The difference lay in how they applied AI rec- ommendations. Low achievers followed generic advice such as doing more adver- tising; high achievers used AI to find tai- lored solutions, such as securing new pow- Pulling up the ladder Impact of generative Al on the gap between high- and low-performing workers ل ا ا Study Topic Inequality Peng et al. (2023) Coding efficiency Brynjolfsson, Li and Raymond (2023) Customer chat Noy and Zhang (2023) Writing quality Dell'Acqua et al. (2023) Product design Chen and Chan (2023) Ad effectiveness Choi, Monahan and Legal analysis ↓ Schwarcz (2023) Otis et al. (2023) Profits and revenue ↑ Roldan-Mones (2024) Debating points 个 Toner-Rodgers (2024) Material discovery ↑ Kim et al. (2024) Investment decisions ↑ Source: The Economist with AI; less sophisticated investors saw gains of 2%. Seasoned investors made bet- ter use of insights from earnings calls such as those concerning R&D spending, share repurchases and operating profit before depreciation and amortisation. As AI reshapes work, new tasks are emerging. Rajeev Rajan of Atlassian, an of- fice-software firm, says that AI tools free up a couple of hours a week for engineers, allowing them to focus on creative work. Junior lawyers spend less time on chores and more with clients. "Really smart peo- ple who may be bored with analysing rou- tine earnings releases will benefit the Work smart Kenya, impact of Al assistant on entrepreneurs* 0=control group, using a non-Al business guide Reported learning from provided tool, %-point change All Low performers High performers 3 -5 0 5 10 15 20 95% confidence Reported changes to business that relate to discounting and advertising, %-point change -40 -20 이 20 40 All Low performers High performers Reported changes to business that are more tailored, % change er sources during blackouts (see chart 3). -25 1 2 3 4 5 6 7 8 9 10 Initial productivity decile *Sample of 1,018 scientists Source: "Artificial intelligence, scientific discovery and innovation", by A. Toner-Rodgers, 2024 In financial decision-making, Alex Kim of the University of Chicago and co-au- thors conducted an experiment where par- ticipants used AI to analyse earnings-call transcripts before allocating $1,000 in a simulated portfolio. Sophisticated inves- tors achieved nearly 10% higher returns -20 이 20 40 60 All Low performers High performers *Sample of 640 Source: "The uneven impact of generative Al on entrepreneurial performance",by N.G. Otis et al., 2023