Anthropic, the artificial intelligence firm behind the widely recognised large language model Claude, says it is developing an early-warning system designed to predict whether AI could eventually disrupt white-collar jobs, according to a report by Axios.
The initiative reflects growing concern among economists and policymakers about the long-term labour market effects of generative AI.
Anthropic which has also been in the headlines because of tensions with the United States Department of Defense after reportedly declining to allow the Pentagon to use Claude in certain contexts. At the same time, Anthropic is focusing on research that examines how advanced AI systems could reshape employment patterns.
As part of that effort, Anthropic has introduced a new index intended to track how different occupations may be affected by automation driven by large language models. The index aims to measure how much of the work performed in specific jobs can potentially be handled by AI systems like Claude.
Initial findings from the project suggest there is currently limited evidence that AI tools are directly causing widespread job losses. However, the company believes building a monitoring system now before major disruption occurs—could help economists and policymakers detect changes in the labour market much earlier.
What did Anthropic economists say?
In a new research paper, Anthropic economists Maxim Massenkoff and Peter McCrory explained that the goal of the index is to identify potential economic shifts before they become clearly visible in employment data.
According to the researchers, establishing this framework early could make future analysis more reliable. They wrote that “by laying this groundwork now, before meaningful effects have emerged, we hope future findings will more reliably identify economic disruption than post-hoc analyses.”
The initiative comes amid an ongoing debate among economists about how best to measure the real-world impact of AI on employment. Meanwhile, Anthropic’s CEO Dario Amodei has been among the most vocal technology leaders warning that increasingly capable AI systems could significantly disrupt parts of the job market, particularly knowledge-based professions.
How does the index work?
Anthropic’s index works by analysing the specific tasks that make up different occupations. Researchers break jobs down into individual activities and estimate which of those tasks could be performed by large language models.
The system also studies anonymised usage data to identify which tasks people are already delegating to AI tools today. By combining task analysis with real-world usage patterns, the index aims to provide a clearer picture of where AI adoption is happening—and where future disruption might occur.