印度策略：AI vs 人类—解码60国AI使用现状

Al vs Human: Decoding Al usage across 60 countries important question is adoption. Ultimately,the extent and quality of Al usage will determinewho captures the resulting economic gains. Our proprietary LLMs vs. Coders study,whichanalysed Alusage acrossglobal techcompanies,offered limited evidencethat adoption hasyet translated into meaningful productivity gains. In this report, based on LLM usage stats,we examine how Al is being adopted globally.The results uncover notable divergences inadoption patterns and raise a broader question:can these earlytrends help identifythefuturewinnersandlosers intheAl era? +65 6326 7643venugopal.garre@bernsteinsg.com +91 226 842 1482nikhil.arela@bernsteinsg.com ChatGPT's launch in late 2022.By March 2026, nearly 18% of the global working-agepopulation was using Al, up from 15% just nine months earlier.The prevailing narrativeis that developed markets are pulling ahead while emerging markets risk falling behind,threatening the labor-arbitrage models on which many economies were built. At firstglance, the data appears to support this view.But per-capita usage is a poormeasure.Emergingmarkets accountfor a disproportionate share of the world's population andinherently less Al-intensive. When viewed through the lens of absolute usage within theinformation economy,the gap narrows dramatically.In fact, several emerging markets rivalor surpass many developed nations. The great usage divide: If the debate on adoption is largely settled, the more importantquestion is how Al is being used. Here, a striking divide emerges. Emerging marketsdeveloped markets lead in sales,finance and healthcare applications.Middle-incomeeconomies also showa greater concentration ofusage in a handfulof domains,whereasdeveloped markets display a broader and more diversified adoption profile. The disproportionategains from Al: This is where the findings become most surprising.greater value from Al. Users in EMs report average time savings of 4.6 hours pertaskversus 3.8hours indeveloped markets.More importantly,16% of tasks delegated to Alarebeyondwhatusers could haveperformedthemselves,compared with12% inhigh-incomeeconomies. The data points to two powerful trends: diminishing productivity returns as Aladoptionmatures,andtheunintended consequences of excessive human oversight, whereworkers gradually lose the ability to effectively review and challenge Al-generated outputs. quality, while EMs devote more than half of Al activity to automation. The implication isuncomfortable:many of the routine,process-driven jobs mostvulnerable to disruptionare already being automated, often within the very EMs that became global hubs for suchare blurring, with weekend Al activity among top-skilled workers rising by 8%. As softwarethe stack. Based on our analysis of the data and trends across over 6O countries (Ex China, as LLMs data are based on US LLMs),we expect the following trends to play out in the medium to long term, as Al adoption scales up and peaks:1. IT services industry will becomeless and less labor-intensive in the future,and much of their competition will emerge from software engineers within the same countries,and consulting firms who have positioned bettertowards techand Al.Thehigh usage in the field of computers,and within that a disproportionate share in automation tasks, is atestamentto that idiosyncracies,plus language constraints.This starts making a strong casefor local LLMstrained in vernacularlanguages,laws andcustoms specifictogeographies andvernacularlanguages andalignedwithlocal professions and norms.Weexpect a further sovereign push to Al, which will be instrumental in hitting the usage cap disrupted by Al (along with healthcareandfinance),with streaming platforms, contentand production houses winning whilelegacybroadcasters andstudios onthereceivingend LLMs,most of which arerunning heavy losses. This may happen even as per-token costs fall. Data center construction playmight peak in the medium term, but the long run benefits will accrue to cloud service providers and power companies. WHAT'SNEWINTHISREPORT?It's importantto address this question first, since we have written quite a lot on Al overthe last two and a half years.For those and strategictool (hereandhere),mediatrends shapedbyAl (here),impactondevelopedvis-a-vis developingmarkets (hereandhere),thelesser-discussedbeneficiariesofAlwave(here),realityofAladoption,P&Limpactandjobdisplacement(herehere and here),future of Al trends (here), emerging Al startups (here and here), capabilities of Al vs Humans (here and here) andlastly,the silent social media impact of Al (here). If you notice, all of these topics center around either of these two. i) the supply side: what's the capability of models, are theyenough to replace humans,whatare the emergingtrends and startups; orii)the sovereign and corporate impact: how much isthe adoption, are nations strategizing Al. We haven't really lookedat the demand sidefrom an end consumer per