Ask Claude to compare two short articles and you'll typically get: an opening acknowledgment, the comparison itself with headers and bullet points, and a closing summary of what was just compared. For a two-minute task, that's reasonable output. For a quick lookup you'd have Googled a few years ago, it's three times the words you need.
This is a design gap in Claude's current setup. The model defaults to thorough - useful when you're debugging code or reviewing a contract, actively annoying when you just want a quick answer. A verbosity setting in Claude.ai's preferences - something like a slider from "concise" to "detailed" - would let users match output length to their actual use case without typing "be brief" at the start of every prompt.
The workaround already exists: add instructions to your custom system prompt, or prepend "briefly" before each question. But that's a workaround, not a solution. Casual users doing quick lookups are the least likely to know custom instructions are even an option, and the most likely to be frustrated by essays in response to simple requests.
ChatGPT and Gemini have the same default-verbose problem. But the mismatch is most noticeable in Claude because it's particularly prone to restating the task before executing it, then summarizing its own output afterward. A simple account-level preference, defaulting to "standard" with a "concise" option, would be a low-effort change that would make the model noticeably better for everyday use.