Follow-up: Core usability pain points from a long-term user perspective
To continue from my earlier point about the limitations of a traditional, 90s-style forum model, I want to clarify my concrete usability pains as a ChatGPT user.
Voice interactions and conversation fragmentation
A core issue is that voice interactions always start a new conversation instead of continuing the active one. This fragments context across multiple threads. What is logically a single discussion ends up split into several conversations, each missing parts of the reasoning and background. This makes it difficult to maintain continuity, revisit decisions, or understand prior context.
Conversation growth and management overhead
Because of this fragmentation, the total number of conversations grows rapidly. Managing them becomes increasingly time-consuming, especially since the current UI is very basic and slow. There are no bulk actions, and conversations cannot be selected or managed efficiently at scale.
Ideally, conversations should be manageable in a structured way, similar to a file system or table view, for example:
-
sorting by name, creation date, last activity date, or conversation length
-
folders and/or tags (labels, hashtags) to group related discussions
-
bulk actions such as archive, delete, or move
Projects help somewhat, but they feel like a partial solution layered on top of a limited underlying system.
Lack of customization and user control
Users currently have almost no ability to customize the UI or UX to fit their workflows. Features and behaviors change frequently, good workflows disappear, and there is no way to opt in, opt out, or preserve preferred usage patterns. Development feels driven by rapid experimentation, while users have no control over stability or continuity.
This makes it difficult to build reliable long-term workflows, because usability depends entirely on what happens to be released next.
Transparency and communication
Another major issue is the lack of communication around changes. New features and behavioral changes often arrive without explanation or advance notice. Users are left guessing why changes were made, what problem they solve, or what the long-term vision is.
Even high-level explanations would help users adapt and evaluate changes more constructively.
User perspective and broader direction
Finally, it’s concerning that the user perspective is not clearly reflected in the company’s stated mission or values. Given how central ChatGPT has become to people’s daily work and thinking, this absence is noticeable.
I’ve also sent similar feedback via the Help Center chat (https://help.openai.com/en), but the core frustration remains: users have lost the ability to adapt the tool to their needs.
For me personally, my productivity system built around ChatGPT was significantly disrupted by recent changes—especially voice creating new conversations and the lack of scalable conversation management. The UI is visually clean, but functionally too limited and slow for advanced, high-volume use.
This feedback comes from appreciation for the product’s potential. ChatGPT is powerful, but the current lack of continuity, control, and predictability makes serious long-term use unnecessarily difficult.
