Anthropic: AI Fluency Is Becoming a Class Gap
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Anthropic's Economic Index report (March 2026) analyzed over 1 million Claude conversations and found users with six or more months of experience achieve a 10% higher success rate than newcomers — a gap that holds across tasks, countries, and model versions
- Automated sales and outreach and automated trading workflows each doubled in prevalence on Claude between November 2025 and February 2026, intensifying displacement risk for workers in those roles
- Anthropic CEO Dario Amodei previously warned that AI could eliminate half of all entry-level white-collar jobs, a forecast that the new data on two-tier usage now contextualizes
- Washington, D.C. shows Claude adoption roughly four times what its population would predict, the report's sharpest example of how geography and education are concentrating AI fluency
- An NBC News poll from earlier this month found 57% of registered voters believe AI's risks outweigh its benefits, with AI's net favorability lower than every topic surveyed except the Democratic Party and Iran — and two points below ICE
- Peter McCrory, Anthropic's head of economics, told Axios the real divide isn't people who use AI versus people who don't, but experienced users versus casual ones, with the report distinguishing low-value "automation" tasks from higher-value "augmentation" work like coaching, stress-testing, and business planning
Why it matters: For workers, the divide isn't about using AI or not — it's about mastery. Anthropic's data shows six-month veterans outperform newcomers by 10% regardless of role, country, or model, meaning casual users face displacement not from AI itself but from AI-fluent colleagues who can offload more work. The two-tier workforce is already visible in the data; neither party has a policy response.
