The $100/Month Controversy

Anthropic quietly changed its claude.com/pricing page, making Claude Code, which had previously been included in the Pro ($20/month) plan, appear to be exclusive to the Max ($100/$200) plan. There was no official announcement, and the confusion grew when the page was later reverted to its original state.

The change immediately spread across Reddit, Hacker News, and Twitter, and Anthropic's Head of Growth explained it as "a small test targeting about 2% of new prosumer signups." However, the author questioned this explanation, pointing out that the exposure was actually broader and that the Internet Archive had also saved a copy of the changed pricing page.

The issue isn't simply a pricing page edit — it's the damage to trust. The gap between $20 and $100 is large, especially for many users, and the way the change appeared to happen without notice created anxiety for both existing and new users.

The author argues that since Claude Code has already defined the coding agent category and is a tool widely used in educational materials and courses, this kind of test was strategically inappropriate. By contrast, they note that OpenAI's Codex continues to be offered on the free and Plus ($20) plans, and said that this incident made Codex look like the safer choice by comparison.

Anthropic subsequently reverted the pricing page, but without a clear official explanation. In a follow-up update, the company stated that the landing page and documentation for logged-out users had been changed for a test, and that it reverted them after they caused confusion for the 98% of users who weren't part of the target group. However, it remains the case that the experiment itself may still be running at the 2% level.