A new survey conducted by Epoch AI and Ipsos in April 2026 has found that Anthropic’s Claude boasts a user base that is significantly more affluent than its major competitors, including ChatGPT and Gemini.
While Claude is often viewed as a “niche” professional tool compared to the mass-market reach of ChatGPT, its penetration among high earners is disproportionately high.
1. The Income Breakdown: Claude’s Affluence Lead
The survey, which analyzed weekly active users in the United States across three waves in March and April 2026, revealed a sharp socioeconomic skew.
| AI Assistant | % of Users in Households Earning $100,000+ |
| Claude | 80% |
| Microsoft Copilot | 64% |
| ChatGPT | 56% |
| Google Gemini | 56% |
| Grok | 56% |
| Meta AI | 37% |
- The Premium Niche: Claude’s concentration of high earners is more than double that of Meta AI. Analysts suggest this is because Claude’s primary use cases—complex coding, long-context document analysis, and high-level reasoning—align closely with “knowledge worker” roles in high-paying sectors like law, finance, and engineering.
- The Meta AI Gap: Meta AI’s lower percentage (37%) reflects its integration into social platforms like WhatsApp and Instagram, which target a massive, broad-based consumer audience rather than a specialized professional one.
2. Relative vs. Absolute Reach
Despite the high concentration of wealthy users, the report clarifies that Claude’s lead is relative, not absolute. When looking at the total number of high-earning individuals using AI, the leaderboard changes.
- ChatGPT Dominance: Among all U.S. households earning over $100,000, 37% use ChatGPT, making it the most used tool even for the wealthy.
- The Rest of the Field: Gemini follows at 24%, Copilot at 14%, and Claude reaches only 6% of this demographic in total.
- The “Non-User” Majority: Notably, 44% of high earners still report using no AI tools at all, highlighting a massive untapped market for high-end productivity assistants.
3. Economic Implications: The “Negotiation Advantage”
The survey results align with a recent Anthropic research paper on the economic impact of AI models.
- Better Outcomes: The study found that users of “stronger” AI models (like Claude 4.6 or GPT-5.5) consistently secured better prices and terms in simulated transactions.
- The Risk of Divergence: Because high earners are disproportionately using the most capable and expensive models, researchers warn of a widening economic gap. Those who can afford (or whose companies provide) premium AI access may gain a structural advantage in negotiations and productivity that further accelerates wealth concentration.