Anthropic launched Claude in 2023. Then it launched Cowork, because Claude was for quick questions and Cowork was for bigger work. Then users started asking which one they were supposed to use for the thing they were trying to do right now. That confusion is, apparently, over.

Starting today, there is one Claude. No mode to select. You describe what you need, and Claude decides whether that is a quick reply, a research session, or a formatted slide deck, then hands it back as a finished file. Claude Docs and Claude Slides are launching in beta alongside the merge, both exportable as PDFs or PowerPoint files. The experience rolls out to Pro and Max plan users over the coming weeks.

The mechanism is intent routing: the unified interface sits on top of a dispatch layer that picks the right tool based on what you describe. This is not a new idea in software, but it is a new idea for Anthropic's product, which until today required the user to make that dispatch call manually. The problem Anthropic heard from users was not the tool itself but the choice of which product a task belonged to, because work started in one did not carry into the other.

The blast radius is narrow today and will expand. Enterprise users who built workflows in Cowork's separate product surface should watch the rollout cadence, because this is a gradual deployment, not a same-day switch. Anyone who has been deep-linking to specific Cowork endpoints or treating it as a distinct agent surface needs to verify how the merge lands in their stack. For most users this is a clean simplification, the kind of product decision that should have shipped earlier.

The pattern: this is Anthropic's third product consolidation of 2026. Claude Design launched as a separate visual product, then got folded in. The API unified under Projects. Now the consumer product surface collapses to one. The direction has been consistent. The execution has been incremental enough that each step felt slightly late.

The read: Anthropic built a second app to solve a problem the first app should have handled, told users for a year which one to use and when, then shipped a merge and called it resolved. The merge is real and the product is better for it. The lesson is that the routing decision was always Anthropic's to make, not the user's, and they made it about twelve months after it became obvious.

The builder's move: if you are on Cowork's API or have hard-coded product surface assumptions, pull your enterprise dashboard now. The rollout is weeks, not days. Claude Docs and Claude Slides are in beta, which means no production commitments yet. Test the new routing behavior before it hits your team's workflows.

The contrast against the rest of the field is the sharpest thing about today. The same morning Anthropic is compressing its product surface to a single point of entry, OpenAI is doing the opposite, layering sponsored agents inside ChatGPT as a product-within-a-product. One lab is betting that simplicity scales. The other is betting that surface area monetizes. These are not compatible theories, and one of them is wrong.

Anthropic framed this as an increase. It is not, relative to what paid users have had since May. The math is simple and the communication was not.

The mechanism: in May, Anthropic added a 50 percent temporary boost to Claude Code's weekly usage allowances for Pro, Max, Team, and seat-based Enterprise plans. It extended the boost in July. Then again in late July. Then in August. Then on August 31. On September 13 the temporary boost expired. On September 14 the permanent 25 percent increase over the original baseline activated. The net result is that users who had 150 percent of the original allowance now have 125 percent, a 17 percent reduction from the level they ran at all summer.

The blast radius lands heaviest on power users, particularly Claude Code subscribers on the $200 per month Max plan who have been running Fable 5.1 for extended agent sessions. Fable 5.1 is computationally expensive. Users posted screenshots of weekly limits consumed in single sessions. Some announced cancellations and migration to OpenAI Codex for less intensive tasks. A class action lawsuit has been filed by Max subscribers, though the legal theory is not yet public.

The pattern is a behavioral economics problem. When you extend a temporary discount for five consecutive months, it stops being temporary in the user's mental model. The boost became the baseline. Anthropic's framing as "permanent increase over original" was accurate as written and misleading as experienced. The five extensions were announced quietly and expired on a date most users had long stopped tracking.

The read: if you extend a temporary benefit long enough for users to budget around it, you own that benefit. Anthropic ran a five-month experiment in expectation-setting and failed the final exam. The underlying policy is defensible. The framing was not.

The builder's move: reset your mental model to 125 percent of the May baseline, not 150. If you are running Fable 5.1 for heavy agent workloads on Max, audit your session lengths now. API access has separate limits, so teams doing production work should evaluate whether a direct API contract is more predictable than the subscription tier for their use case.