Anthropic ran a silent 5x price test on Claude Code and reversed it within hours

Anthropic silently updated its claude.com/pricing page on April 22 to restrict Claude Code access to its $100/month and $200/month Max plans—a 5× increase over the $20/month Pro tier where the feature had previously lived—then reversed the change within hours after backlash on Reddit, Hacker News, and Twitter. No formal announcement accompanied either move.

Anthropic removed the Claude Code checkbox from the Pro column in the pricing grid, with no blog post, changelog, or email to existing subscribers. The Internet Archive captured the updated page before it reverted, corroborating the screenshots that initially circulated. Amol Avasare, Anthropic's Head of Growth, offered the only near-official response via tweet, describing the change as "a small test on ~2% of new prosumer signups" and noting that existing Pro and Max subscribers were not affected. Anthropic has still not issued a formal statement as of publication.

Developer and AI educator Simon Willison flagged the change in real time and challenged the 2% framing, noting that "everyone I've talked to is seeing the new pricing grid" and that the Internet Archive had already captured a copy. He also pointed out that Claude Cowork—described as "effectively a rebranded version of Claude Code wearing a less threatening hat"—remained available on the $20/month plan throughout the episode.

The episode surfaces a concrete vendor-trust risk for enterprise teams evaluating AI coding toolchains. Willison put it plainly: "Strategically, should I be taking a bet on Claude Code if I know that they might 5x the minimum price of the product?" The concern is amplified by Claude Code's market position—Willison credits it with being responsible for "billions of dollars in annual revenue for Anthropic already" and for defining the coding-agent category outright.

OpenAI moved quickly to exploit the opening. Thibault Sottiaux, engineering lead for OpenAI Codex, posted: "I don't know what they are doing over there, but Codex will continue to be available both in the FREE and PLUS ($20) plans. We have the compute and efficient models to support it. For important changes, we will engage with the community well ahead of making them. Transparency and trust are two principles we will not break, even if it means momentarily earning less." The statement is a direct pitch to developers weighing which coding agent to build workflows around.

The transparency failure is more damaging than the pricing test itself. A silent grid change—discovered by screenshot, confirmed by an employee tweet, reversed with no explanation—gives CTOs and procurement teams no basis for contractual or roadmap planning. If a $20-to-$100 move can appear and disappear in an afternoon, any current Pro-tier pricing commitment carries implicit optionality for Anthropic. Willison, who has published 105 posts teaching Claude Code and ran a Claude Code tutorial at the NICAR data journalism conference last month, said the incident has shaken his trust in Anthropic's pricing transparency and that he is actively reconsidering Codex as his default teaching tool.

What remains unresolved: Anthropic has not committed publicly to keeping Claude Code on the $20/month tier, has not explained why the test launched without any user communication, and has not addressed the Claude Cowork anomaly—a near-identical product that stayed available at Pro pricing throughout. Until Anthropic makes an explicit, durable commitment on Claude Code's tier floor, any enterprise build-out that treats the $20 entry point as stable is carrying undisclosed pricing risk.