#Advance Voice Mode: Canvas Document Issue

1 messages · Page 1 of 1 (latest)

cursive craterBOT
#

Reported by @surreal geyser

Bug Report: Advance Voice Mode: Canvas Document Issue
`Steps to Reproduce`
  1. Initiate Advanced Voice Mode on smartphone.
  2. Start screen sharing with ChatGPT.
  3. Ask ChatGPT to create a new Canvas document using the document/block of text on the screen.
`Expected Result`

ChatGPT creates a new Canvas document and updates the document as we work together via the screen share and voice interaction.

`Actual Result`

ChatGPT will say that it will create a Canvas document and to give it a second or minute. It sounds like all is working, but when you go back to the chat, no Canvas document was created whatsoever. All the while, ChatGPT carries on the conversation with you telling you it is writing/copying down what you two are working on with no result.

`Environment`

iOS 26.1 | iPadOS 26.1

#
Additional Information

Please provide relevant details to help resolve the issue, such as:

  • ChatGPT Shared Link (if applicable).
  • Screenshots or videos demonstrating the problem.

-# ➜ Need to contact support? Visit the OpenAI Help Center.

ionic dagger
#

I think the best solution to this is to add tool awareness to your AI OS. When it learned to detect its tools, learns its constraints, and then learns to be transparent about constraints. It starts to act differently.