Anthropic tested removing Claude Code from the Pro plan
A 2% pricing test updated public docs, causing confusion before a swift reversal.
Anthropic sparked a developer uproar after its public pricing page indicated Claude Code, the popular agentic development tool, would no longer be included in the $20/month Pro subscription, reserving it for the $100+ Max tier. New users reported being unable to access the feature, while existing subscribers saw no change. The confusion stemmed from Anthropic running a small-scale test on approximately 2% of new 'prosumer' signups but updating its official documentation as if the change was universal, leading to widespread speculation and frustration on social media.
Head of Growth Amol Avasare explained the rationale, stating that user engagement has fundamentally changed since the Max plan launched a year ago. The rise of tools like OpenClaw, the introduction of Claude Opus 4 and Cowork, and the shift to 'always-on, multi-agent workflows' have dramatically increased compute consumption per user. This test was part of exploring options to manage infrastructure strain, which has also led to peak-hour limits and occasional outages. Following the backlash, Anthropic reverted its pricing page and Avasare assured users that any future plan changes affecting existing subscribers would come with ample advance notice directly from the company.
- Anthropic tested removing Claude Code from the $20 Pro plan for ~2% of new signups, but updated public docs universally.
- The test was driven by explosive growth and a shift to heavy, always-on agent workflows consuming significant compute.
- After swift user backlash on Reddit and X, the company reverted the pricing page and promised clear communication for any future changes.
Why It Matters
Highlights the infrastructure strain from AI agent adoption and the delicate balance between pricing, features, and user trust.