News for hackers. Stuff that matters.

A comment on ChatGPT Developer Mode. Parent comment

Re:ChatGPT Developer Mode by simonw on Wednesday September 10, 2025 @10:10PM (#45206890)
If you have an MCP tool that can perform write actions and you use it in a context where an attacker may be able to sneak their own instructions into the model (classic prompt injection) that attacker can make that MCP tool do anything they want.
[ Parent | on HN ]
Re:ChatGPT Developer Mode by CGamesPlay on Wednesday September 10, 2025 @10:45PM (#45207126)
How is this "developer mode" different than just connecting the MCP normally and prompt injecting it to use the same tools?
[ Parent | on HN ]
Re:ChatGPT Developer Mode by simonw on Wednesday September 10, 2025 @11:27PM (#45207418)
It's no different. This just brings that unsafe anti-pattern to the ChatGPT consumer app itself - albeit hidden behind an option with a scary name that might hopefully discourage many users who don't understand the consequences from turning it on.
[ Parent | on HN ]