OpenAI will now also display an Elevated Risk label when you access certain features that could be risky.



Source: ADnet
Source Link: https://www.zdnet.com/article/how-chatgpt-lockdown-mode-protects-prompt-injection-attacks/