AiNews.com
Posts
Anthropic Launches Economic Index to Track AI’s Impact on Jobs

Anthropic Launches Economic Index to Track AI’s Impact on Jobs

Alicia Shapiro
February 11, 2025 • Estimated Reading Time: 8 minutes

A modern infographic-style illustration featuring professionals from various industries—software developers, writers, educators, and office administrators—working alongside AI tools. Augmentation is depicted with AI assisting humans in tasks like coding and content creation, while automation shows AI completing tasks independently. The background integrates subtle charts and graphs, highlighting data on AI’s impact in the workforce. The design is sleek, emphasizing the evolving relationship between AI and human work.

Image Source: ChatGPT-4o

Anthropic Launches Economic Index to Track AI’s Impact on Jobs

Anthropic has unveiled the Anthropic Economic Index, a new initiative aimed at tracking the evolving impact of AI on labor markets and the economy. The index’s inaugural report provides a detailed analysis of how AI tools, particularly Claude.ai, are being integrated into real-world tasks across various industries. This initiative also includes an open-source dataset, inviting researchers, economists, and policy experts to contribute insights and help shape future studies.

Key Findings: AI Augments More Than It Automates

Drawing on millions of anonymized conversations from Claude.ai, the initial report highlights several key trends in AI’s integration into the workforce:

AI Augments More Than It Automates: About 57% of AI usage involves augmentation, where AI enhances human capabilities, while 43% involves automation, where AI directly performs tasks.
AI Is Concentrated in Tech and Writing: The highest usage of AI is found in software development and technical writing. Over 36% of occupations use AI for at least a quarter of their tasks, while only 4% of jobs rely on AI for more than three-quarters of tasks.
Mid-to-High Wage Jobs See More AI Use: AI is most commonly used in mid-to-high-wage occupations like computer programming and data science. Conversely, both the lowest- and highest-paid roles—such as manual laborers and specialized medical professionals—show limited AI adoption.

A horizontal bar chart comparing AI usage by job type based on conversations with Claude versus the percentage of U.S. workers in each occupation. Orange bars represent the percentage of Claude conversations, while gray bars show workforce representation. The Computer and Mathematical category has the highest AI usage at 37.2%, despite only comprising 3.4% of U.S. workers. Other notable disparities include Arts and Media (10.3% AI usage vs. 1.4% workers) and Office and Administrative Support (7.9% AI usage vs. 12.2% workers). Farming, Fishing, and Forestry shows the lowest AI usage at 0.1%.

AI Usage by Job Type: Comparing Claude Conversations to U.S. Workforce. Image Source: Anthropic

Mapping AI Across the Modern Economy

The report reveals that computer and mathematical occupations account for the largest share of AI usage, representing 37.2% of Claude.ai queries. This includes tasks like code debugging, software modification, and network troubleshooting.

Other notable categories include:

Arts, Design, and Media: Making up 10.3% of AI usage, largely driven by writing, editing, and creative tasks.
Education and Library Services: Representing 9.3% of usage, with AI assisting in lesson planning, content creation, and research.
Office and Administrative Support: Comprising 7.9% of queries, where AI is used for tasks like scheduling, data entry, and document formatting.

In contrast, industries requiring significant physical labor, such as farming, fishing, and forestry, showed minimal AI integration, accounting for just 0.1% of usage.

A colorful infographic highlighting six job categories with the highest AI usage based on Claude.ai data. Computer & Mathematical (37.2%): Includes Computer Programmers and tasks like software development and debugging. Arts & Media (10.3%): Covers Technical Writers and tasks like editing and public relations management. Education & Library (9.3%): Features Tutors and tasks like curriculum development and teaching. Office & Administrative (7.9%): Includes Bioinformatics Technicians and IT system administration. Life, Physical, & Social Science (6.4%): Involves Clinical Psychologists and academic research. Business & Financial (5.9%): Highlights Security Management Specialists and financial analysis tasks.

Top AI-Integrated Job Categories and Tasks from Claude.ai. Image Source: Anthropic

Automation vs. Augmentation: What’s the Difference?

The Index distinguishes between tasks where AI automates versus augments human efforts:

Augmentation (57%): AI collaborates with users, helping with brainstorming, skill development, or validating work. For example, Claude might assist in drafting content, which the user then edits and refines.
Automation (43%): AI fully handles tasks, such as generating reports or formatting documents without further user input.

Interestingly, many tasks that appear automated in data could involve behind-the-scenes human oversight, blurring the line between augmentation and automation.

A stacked bar chart comparing the percentage of Claude conversations involving augmentation versus automation. Augmentation accounts for 57.4% of conversations and includes Validation (2.8%), Task Iteration (31.3%), and Learning (23.3%). Automation makes up 42.6%, broken down into Feedback Loop (14.8%) and Directive (27.8%). The chart highlights that AI is more frequently used to augment human work than to fully automate tasks.

Augmentation vs. Automation: How AI Assists and Automates Tasks. Image Source: Anthropic

AI Use and Wages: A Complex Relationship

The report also explores how AI usage correlates with wages. Mid-range salaried professionals, such as computer programmers and copywriters, are the heaviest users of AI tools. Conversely, low-wage jobs (like shampooers) and high-wage roles (such as obstetricians) show minimal AI involvement. This reflects both the current limits of AI capabilities and practical barriers in specific fields.

A scatter plot comparing median annual wages (x-axis) to the percentage of Claude conversations (y-axis) for various occupations. Computer Programmers show the highest AI usage at over 6% with salaries around $75,000. Software Developers and Copywriters also have high AI engagement. Lower-wage jobs like Shampooers ($25,000) and high-wage roles like Obstetricians ($200,000) show minimal AI usage. A vertical dotted line marks the U.S. median wage of $60,070.

AI Usage and Wage Correlation Across Occupations. Image Source: Anthropic

Open Data and Future Research

Anthropic’s decision to open-source its dataset marks a significant step toward transparency and collaborative research. The data, drawn from approximately one million anonymized conversations on Claude.ai, allows researchers to explore AI’s effects on labor markets in greater depth.

Anthropic’s Clio system was used to map these conversations to occupational tasks based on the U.S. Department of Labor’s O*NET classification. This methodology provides a nuanced look at how AI is affecting specific tasks within broader occupations.

The Index will be updated regularly to track changes in AI adoption and its long-term economic impacts. Anthropic invites economists, policymakers, and researchers to provide feedback and suggest new areas of inquiry.

For more details, visit Anthropic’s blog post.

What This Means

The Anthropic Economic Index offers an unprecedented look at how AI is being integrated into various sectors of the economy. While fears of widespread job automation persist, the data suggests that AI is currently more focused on augmenting human work rather than fully replacing it.

However, as AI models become more advanced, the balance between augmentation and automation may shift. Understanding these dynamics is crucial for policymakers and businesses as they prepare for the evolving labor landscape. By openly sharing their data and inviting collaboration, Anthropic is positioning the Index as a key resource for tracking AI’s role in shaping the future of work.

Editor’s Note: This article was created by Alicia Shapiro, CMO of AiNews.com, with writing, image, and idea-generation support from ChatGPT, an AI assistant. However, the final perspective and editorial choices are solely Alicia Shapiro’s. Special thanks to ChatGPT for assistance with research and editorial support in crafting this article.