This post is part of the AI design patterns series – I’m sharing new design patterns I’m seeing as AI enables solutions that weren’t possible before.
Writers create their first drafts with ChatGPT. Dentists are building applications with Claude. Engineers fix bugs and build brand new tools faster with Cursor, etc.
Making digital things from scratch isn’t the norm anymore. AI does the heavy-lifting of creation – a bulk of the work now goes into reviewing and editing the artifacts AI creates to meet our standards.
If your product allows users to create “things”, asking users to start from a blank slate is an outrageous user expectation in 2025. Instead, can your product create those things upfront and just ask users to review and approve?
Intercom Fin doesn’t just create new help center articles from scratch – the product also surfaces edits that can be made to existing articles. Fyxer keeps email responses ready when you open Outlook or Gmail, for almost all your emails.
In both these cases, the user needs to just tweak (only if needed) and mostly approve. No more writing articles or emails from scratch.
So, the more you reduce creation effort and inputs, the more you increase the chances of user delight 🙂
This post is part of the AI design patterns series – I’m sharing new design patterns I’m seeing as AI enables solutions that weren’t possible before.
Shankar is thinking 