Survey finds Claude's weekly active users in the US skew far wealthier than any rival AI assistant

A new survey reveals Claude's US user base earns substantially more than ChatGPT, Gemini, and other AI assistant competitors, suggesting wealth concentration among premium AI adopters and potential market segmentation by income level.
Modelwire context
Analyst takeThe more pointed finding isn't just that Claude skews wealthy — it's that this income skew is apparently sharper than any rival, which suggests Anthropic may have inadvertently (or deliberately) built a product whose natural ceiling is a narrow, high-income demographic rather than mass adoption.
This is largely disconnected from recent activity in our archive, as we have no prior coverage to anchor it to. But it belongs to a broader conversation about whether the AI assistant market is fracturing into distinct tiers: a premium segment willing to pay for Claude's more careful outputs, and a volume segment served by ChatGPT and Gemini at lower or no cost. That split has real consequences for Anthropic's revenue model, since a smaller, wealthier user base can sustain subscription revenue but limits the network-effect and data-flywheel advantages that come with scale. For Google and OpenAI, the same data is arguably good news: their user bases appear broader, which matters for enterprise sales narratives and advertiser interest alike.
Watch whether Anthropic responds to this perception by introducing a lower-cost or free tier within the next two quarters. If they don't, it signals they are consciously trading breadth for margin, and the income skew will likely widen further.
This analysis is generated by Modelwire’s editorial layer from our archive and the summary above. It is not a substitute for the original reporting. How we write it.
MentionsClaude · ChatGPT · Gemini · Anthropic · OpenAI · Google
Modelwire summarizes — we don’t republish. The full article lives on the-decoder.com. If you’re a publisher and want a different summarization policy for your work, see our takedown page.