[ChatGPT Web] Add Explain / Translate popovers for selected text without affecting the main conversation
Nobody has claimed this yet.
- Dominant language
- Rust
- Stars
- 125k
- Forks
- 19.4k
- PR merge metrics
- PR metrics pending
Description
What feature would you like to see?
Please add Explain and Translate actions to the selected-text toolbar in ordinary ChatGPT conversations, with the result shown in a lightweight popover / overlay instead of being sent as a new message to the current chat.
The core requirement is: temporary clarification should not modify the main conversation context.
Current behavior
When I am reading a ChatGPT response and encounter an unfamiliar technical term, abbreviation, or foreign-language phrase, I can select it and use the existing selected-text action to add it to the chat / ask about it.
That works for genuine follow-up questions, but it is unnecessarily disruptive when I only want a quick definition or translation.
For example, I may be in the middle of a long discussion about sales strategy and encounter a term such as ICP. I only need a 10-second explanation of the term before continuing to read the original answer.
Today, asking about it through the main chat has several costs:
- it inserts an unrelated side question into the main transcript;
- it interrupts the user's reading and reasoning flow;
- the temporary question becomes part of the conversation context;
- the assistant may incorporate that side question into later turns;
- long conversations accumulate many low-value clarification messages that make the transcript harder to review.
Proposed interaction
When text is selected in an assistant response, add actions such as:
- Explain
- Translate
Clicking either action should open a small contextual popover anchored to the selection.
Example:
Selected text:
ICPExplain
ICP (Ideal Customer Profile)— the profile of the type of customer that is the best fit for a product or service, typically considering likelihood to buy, value received, and long-term value.
Closing the popover should return the user to exactly the same reading position.
No new chat message should be created, and the main conversation context should remain unchanged.
If the user wants to continue discussing the explanation, the popover can provide an optional Add to chat / Ask follow-up action.
Suggested behavior
Explain
- Use the surrounding response as local context so the explanation matches the meaning intended in that passage.
- Prefer a short explanation by default.
- Optionally allow
More detailfor a longer explanation.
Translate
- Translate the selected text directly in the popover.
- Infer the target language from the user's ChatGPT language/preferences when reasonable.
- Keep an optional action to continue the translation discussion in the main chat.
Reference interaction
Some AI reading/chat products, such as Doubao, already provide this pattern: selecting text exposes Explain and Translate actions that show the result in an overlay instead of injecting a new turn into the primary conversation.
This works particularly well for long-form reading and technical discussions because the user can resolve a local comprehension problem without changing the thread.
Why this matters
ChatGPT is increasingly used not just for short Q&A but for long-form reading, research, learning, technical work, and extended decision-making conversations.
In those workflows there is an important distinction between:
- a real follow-up that should become part of the conversation, and
- a transient comprehension aid that should disappear after the user understands the text.
The current selected-text workflow treats both as conversation turns.
A non-conversational Explain / Translate popover would reduce context pollution, preserve the main topic, and make long responses much easier to read interactively.
Acceptance criteria
- Selecting text in a ChatGPT response exposes Explain and Translate actions.
- Either action can return a result without sending a new user message.
- The result appears in a temporary popover / overlay near the selection.
- Closing the popover leaves the main transcript unchanged.
- The temporary explanation/translation is not added to the main conversation context by default.
- Users can explicitly promote the result into the main chat if they want to continue discussing it.
- Explain can use the surrounding passage to disambiguate technical terms.
Related issues
There are adjacent requests in the same tracker, but I could not find an exact ChatGPT Web request for this interaction:
- #39556 — inline
Look Upfor selected text in the IDE extension. - #21924 —
Translateaction for selected text in Codex Desktop. - #37520 — restoring
Ask in side chatfor selected text in the Codex app.
This request is specifically for ordinary ChatGPT reading/chat UX and for a zero-main-context-change Explain / Translate interaction rather than another side conversation.
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at the selected-text toolbar in the ChatGPT Web reading flow and trace the existing action that adds selected text to the main chat. Done means Explain and Translate open temporary contextual results, leave the transcript and context unchanged, preserve the reading position, and offer an optional way to add the result to chat.
Written by the indexing model from the issue text.
Assessment
- Domain
- frontend
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Active
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100