Claude Creator Anthropic Reveals the Jobs Most Likely To Use AI

On Tuesday, AI giant Anthropicreleased a new report offering insights into the impact of AI on the U.S. economy, as well as the kinds of people making use of new large language models.

The company's new Anthropic Economic Index shows that technical jobs are the largest users of the company's AI models, including Claude and Claude Code.

Chief among the fields utilizing the models are Computer and Mathematical jobs, including software developers, IT professionals, and data-focused knowledge workers. Cumulatively, they represent about 7% of the company's usage.

However, technical jobs aren't the only ones supplementing their jobs with help from the new machines.

Jobs in the Arts, Design, Entertainment, Sports, and Media represented 5% of Claude usage. A smaller share was used by workers in Community and Social Service (2.2%), Office and Administrative Support (1.8%), Educational Instruction and Library (1.1%), and Farming, Fish, and Forestry (0.8%).

The company singles out tasks that Claude has “mostly automated” or “mostly augmented” in various role, which can be examined from role-to-role. Still, there remains a lot of work for Claude to achieve parity with human workers.

“AI is being used to fully (or ”directively“) automate more and more tasks over time,” Anthropic highlights in its study. “We saw directive automation jump from 27% to 39% of all conversations.” That figure was even higher for enterprise customers.

Even then, the vast majority of the 974 occupations tracked by Anthropic are almost entirely, “Tasks that don't appear in [their] data” at all.

Among the more 'knowledge worker' roles accounting for near-zero usage of Claude are Logistics Analysts, Compensation and Benefits Managers, and Insurance Adjusters, Examiners, and Investigators.

And, as you'd expect, a large portion of construction, production, and health care jobs haven't really been touched by the models' creep into knowledge work.

Select states seem to be adopting Anthropic more, as well. The report's “Usage Index” found that users in Washington, D.C, Utah, and California were among the most ardent users of its models.

Recently raising at a $183 billion valuation, Anthropic is one of the world's most-valuable AI startups. It's only dwarfed by the valuation of competitor OpenAI in the private market, which raised at a $300 billion valuation in recent months and also released a similar report on users' usage this week.

Leaders at both firms have sworn by a paradigm shift that will see AI models generate greater productivity, and eventually supplant the need for human knowledge workers.

At the same time, the firms are fighting claims that their models violate copyright law. Anthropic recently settled a $1.5 billion claim that it pirated books to train its models, making it the “largest U.S. copyright infringement settlement ever.” Similar claims are going ahead against other AI companies.