スタンフォード報告書によると、AIが「歴史的な速さ」で普及

AI's rapid acceleration, especially in affluent countries, is among the chief findings of Stanford University's annual AI Index Report.This year's version, published on April 13, is Stanford's most comprehensive yet, offering a data-driven overview of the current state of AI, where the technology is headed, and its implications for society.Compiled by a steering committee of academic and industry experts and produced by Stanford's Institute for Human-Centered AI, the report was first published in 2017, and this year stretches to an extensive 423 pages. It is based on large-scale surveys and expert analysis.The key finding, AI's rapid spread, was that 53% of the world's population now use generative AI -- faster uptake than the personal computer or the internet.However, the report also acknowledged a global imbalance, with high usage rates in countries strongly correlating with GDP per capita. This is reflected by the pace of adoption in countries such as Singapore (61%) and the UAE (54%), while the U.S. ranked 24th globally at 28.3%.Related:The Real AI Shift Isn’t New Models. It’s Control.In tandem with this, the survey noted that public sentiment toward AI is "growing more complex." While a majority of people now feel optimistic about the benefits of AI, nervousness about the technology has also risen to about 52% of those surveyed for the report.That finding was also dependent on location, with people in the U.S. in particular expressing specific fears. Only a third of Americans expect AI to improve their jobs -- below the global average of 40% -- and among the countries surveyed, none had lower trust in their government to regulate the tech than the U.S.Also notable was the vast disparity between AI experts and the general public in the U.S. on the potential societal impact of AI. While nearly three-quarters of experts think AI will have a positive impact on how people do their jobs, this sentiment is not reflected in the population at large, where fewer than a quarter agreed with this sentiment -- suggesting that big AI vendors need to do significant work in selling the benefits of the technology.Also worrisome for the U.S. is that its traditional lead in the AI sector -- in terms of model size, performance and research -- has now been erased by China. The report notes that since DeepSeek-R1 was launched in May 2025, the performance of the two countries' top models has been much closer than previously, and as of March this year, Anthropic's top model -- Claude Opus 4.6 -- outperformed China's best by only 2.7%.Related:OpenAI GPT-5.4-Cyber is More Open Than Claude MythosThe gap between the two countries in spending on AI is much wider, however, with U.S. corporate investment reaching $285.9 billion in 2025 -- 23 times more than in China. However, the report noted China provides significantly more state funding for AI development than the U.S.Other areas covered in detail are governance, AI's affect on medicine and science, energy usage and how education can benefit from the AI rollout. About the AuthorContributing WriterGraham Hope has worked in automotive journalism in the U.K. for 26 years, including spells as editor of leading consumer news website and weekly Auto Express and respected buying guide CarBuyer.