OpenAI is testing a big upgrade for ChatGPT’s temporary chat feature. The update will allow you to retain personalization in temporary chat, and still block temporary chat from influencing your account.
For those unaware, Temporary Chat in ChatGPT is a mode where you start with a fresh slate.
In this chat, ChatGPT won’t use or reference your past conversations or any saved memories, even if you normally have memory turned on. It will still follow your Custom Instructions if those are enabled, but the conversation itself is treated as “don’t remember this.”
Source: BleepingComputer
Temporary Chats also won’t show up in your chat history, and they won’t be used to improve the models.
As spotted on X, after a new update, ChatGPT’s temporary chat now retains personalization (memory, chat history, style, and tone preferences).
The chat remains temporary, but it will now have access to your personalization, and it can be turned off.
For safety reasons, OpenAI may still keep a copy of the chat for up to 30 days.
To start one, open a new chat and click the pill-shaped “Temporary” button in the top-right corner.
Other ChatGPT improvements
More recently, OpenAI rolled out the ChatGPT Age prediction model to identify a user’s age based on their chat pattern, conversation behaviour, among other factors.
OpenAI warns that its AI model is not perfect, so it may detect that some adult users are teenagers and restrict their accounts. At the same time, it might be tricked by teenagers with adult behavior.
Restricted accounts can still learn, create, and ask questions, but you won’t be able to interact on topics like violence, gore, viral challenges on TikTok-like platforms, etc.
If incorrectly restricted, adult users can verify their age to unlock the full GPT experience.
As MCP (Model Context Protocol) becomes the standard for connecting LLMs to tools and data, security teams are moving fast to keep these new services safe.
This free cheat sheet outlines 7 best practices you can start using today.