I find it hard to believe that the UX and UI of ChatGPT are still designed this poorly. I’m not the only one—many others in my professional network share the same frustration. One of the most fundamental issues is the input experience when dealing with longer texts. After reading through a large block of content, you’re expected to write a prompt, but the input area gives you no spatial orientation—you never quite know where you’re typing.
There are numerous usability improvements that could and should be made. For example, separating the prompt interface from the input field would drastically improve clarity. There are many scenarios where I want to reference multiple pieces of content in parallel, and from a UX perspective, having the option for multiple, independent input fields—each of which can be tagged or referenced—would be far more intuitive.
Also, why is the input area so small on desktop? It’s an unnatural constraint that makes no sense given the available screen real estate. I’d expect a collapsible or dockable panel system, where users can manage input, history, and settings through a more modular interface. It would reduce cognitive load and create a much calmer, more organized experience.