- Anthropic introduced "observed exposure" to assess AI displacement risk in jobs
- Top 10 exposed jobs include programmers, customer service, and data entry roles
- No major rise in unemployment seen since late 2022 in high-exposure occupations
Anthropic has introduced a new measure called “observed exposure” to assess AI displacement risk in jobs. This tool combines theoretical capabilities of large language models with real-world usage data from Claude — Anthropic's AI — to offer an early-warning system for white-collar job impacts, even though current evidence of widespread AI-induced job loss remains limited.
The metric evaluates how much of an occupation's core tasks are both feasible for AI tools and actually being automated today. According to Anthropic, a job's exposure is higher if:
- Its tasks can be theoretically be automated by AI.
- Significant real-world usage in professional settings.
- Higher proportion of automated patterns.
- AI-impacted tasks make up a larger share of the overall role.
10 Jobs Most Vulnerable To AI, As Per Anthropic
Computer programmers, with 75% task coverage, customer service representatives, and data entry keyers are among the most exposed jobs, according to findings from the new index.
The top 10 most exposed occupations, based on Anthropic's analysis, are:
- Computer programmers
- Customer service representatives
- Data entry keyers
- Medical record specialists
- Market research analysts and marketing specialists
- Sales representatives, wholesale and manufacturing, except technical and scientific products
- Financial and investment analysts
- Software quality assurance analysts and testers
- Information security analysts
- Computer user support specialists
Additional Findings
Despite high exposure in these roles, Anthropic's report shows that no significant overall rise in unemployment has appeared since late 2022 compared to less-exposed jobs. However, it notes suggestive evidence of slower hiring for younger workers (ages 22-25) in exposed occupations, potentially affecting entry-level opportunities.
Also, workers in highly exposed roles tend to be older, female, more educated, and higher-paid. Occupations involving physical labour (e.g., construction, repair) show lower risk.
The research also noted that the gap between current observed exposure and full potential impact remains large, raising concerns about future job disruptions if AI adoption rises.
Also read: OpenAI Releases ChatGPT GPT-5.4 With 1 Million Token Context Window — Key New Capabilities
Essential Business Intelligence, Continuous LIVE TV, Sharp Market Insights, Practical Personal Finance Advice and Latest Stories — On NDTV Profit.
