Output truncation bug in structured responses (tables and enumerated lists)

Title: Output truncation bug in structured responses (tables and enumerated lists)

Description: I’ve repeatedly encountered a bug where the final part of a structured response (especially tables, JSON objects, or “…”-based sentences) is cut off or missing.

Examples:

In tables, the final cell (e.g., row 10, column 5) is often blank or missing.

In sentences like “Another example is…”, the phrase after “…” is frequently missing.

However, numbered points like 1, 2, 3, 4, 5 seem to work normally.

Device/Platform:

Platform: ChatGPT mobile app and web

Device: High-performance smartphone (16GB RAM, 3nm Qualcomm SoC)

Model: GPT-4 (likely 4-o based on behavior)

Request:

Please investigate possible truncation or rendering issues in the current output system.

A feedback channel should also be restored to the mobile app for beta users and testers.

Thank you! I’m happy to provide more examples or help debug.