{"id":401,"date":"2026-03-18T23:33:32","date_gmt":"2026-03-19T02:33:32","guid":{"rendered":"https:\/\/rnlsn.cc\/?p=401"},"modified":"2026-03-20T14:58:19","modified_gmt":"2026-03-20T17:58:19","slug":"labor-market-impacts-of-ai-a-new-measure-and-early-evidence","status":"publish","type":"post","link":"https:\/\/rnlsn.cc\/en\/2026\/03\/labor-market-impacts-of-ai-a-new-measure-and-early-evidence\/","title":{"rendered":"Labor market impacts of AI: A new measure and early evidence"},"content":{"rendered":"\n

Most research on AI and jobs asks what AI could<\/em> do to employment. This study asks what it’s actually<\/em> doing \u2014 right now. That distinction matters more than it might seem.<\/p>\n\n\n\n

Anthropic researchers Maxim Massenkoff and Peter McCrory developed a new metric called observed exposure<\/strong>, which combines theoretical AI capability with real-world usage data from Claude. A task only counts as covered if it’s both feasible for an LLM and<\/em> already appearing in professional, automated workflows.<\/p>\n\n\n\n

The findings are grounding. Even in Computer & Math \u2014 where 94% of tasks are theoretically feasible for AI \u2014 Claude currently covers only 33% of them in practice. The gap between potential and reality is wide across every occupational category.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

The most exposed occupations today include computer programmers (74.5%), customer service representatives (70.1%), and data entry keyers (67.1%). Notably, these aren’t low-wage roles. Workers in the most exposed occupations tend to be more educated, better paid, and more likely to be female \u2014 earning about 47% more per hour than workers with zero AI exposure.<\/p>\n\n\n\n

As for actual unemployment: no significant impact has been detected since ChatGPT’s release in late 2022. The effect on exposed workers is statistically indistinguishable from zero.<\/p>\n\n\n\n

There is one early signal worth watching. Hiring of workers aged 22\u201325 into AI-exposed occupations appears to have slowed by around 14% in the post-ChatGPT era \u2014 a result that is barely statistically significant, but echoes findings from other recent studies.<\/p>\n\n\n\n

What makes this research valuable isn’t just the data \u2014 it’s the framework. By establishing a clear methodology before<\/em> major disruptions emerge, it gives future analysts a way to separate genuine AI-driven effects from ordinary economic noise.<\/p>\n\n\n\n

The clock isn’t ticking toward apocalypse. But it is ticking. And now we have a way to watch it in real time.<\/p>\n\n\n\n

\n\n\n\n

Source: Massenkoff, M. & McCrory, P. (2026). Labor market impacts of AI: A new measure and early evidence.<\/em><\/p>\n\n\n\n

Anthropic. anthropic.com\/research\/labor-market-impacts<\/a><\/em><\/p>\n","protected":false},"excerpt":{"rendered":"

A new Anthropic study introduces a smarter way to measure how AI is actually affecting jobs \u2014 not just what it could theoretically do. The early evidence is clear: no wave of unemployment yet, but the gap between AI’s potential and its real-world impact is closing fast.<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[34,1],"tags":[35,10],"class_list":["post-401","post","type-post","status-publish","format-standard","hentry","category-ai","category-rnlsncc","tag-ai","tag-rnlsncc"],"_links":{"self":[{"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/posts\/401","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/comments?post=401"}],"version-history":[{"count":6,"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/posts\/401\/revisions"}],"predecessor-version":[{"id":422,"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/posts\/401\/revisions\/422"}],"wp:attachment":[{"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/media?parent=401"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/categories?post=401"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/rnlsn.cc\/en\/wp-json\/wp\/v2\/tags?post=401"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}