Microsoft Research has released a peer-review-pending paper that ranks 80 occupations by how closely their daily tasks align with the capabilities of large-language-model chatbots. The study analysed 200,000 anonymised U.S. user exchanges with Microsoft’s Copilot and assigned each job an “AI applicability score” that measures the overlap between chatbot abilities and job requirements. Interpreters and translators, historians, sales representatives of services, writers and authors, and customer-service representatives top the list of 40 roles deemed most exposed to current generative-AI tools. At the other end of the spectrum, largely manual or equipment-intensive roles such as dredge operators, bridge and lock tenders, water-treatment-plant operators, roofers and nursing assistants rank among the 40 least exposed occupations. The researchers stress that high applicability does not guarantee automation. “Our study shows AI supports many tasks—particularly those involving research, writing and communication—but does not indicate it can fully perform any single occupation,” senior researcher Kiran Tomlinson said in the paper. The findings arrive as generative-AI adoption accelerates: workforce tracker Challenger, Gray & Christmas attributed more than 10,000 U.S. job cuts in July alone to the technology.