ChatGPT's new Lockdown Mode can stop prompt injection - here's how it works

23:51 - 17 Feb 2026

www.zdnet.com

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

Article info:

Full Article URL: https://www.zdnet.com/article/how-chatgpt-lockdown-mode-protects-prompt-injection-attacks/
Textarea URL (this page):
https://textarea.nl/i/eNoVyUEKwCAMBdETxex7G4lSrZqIfgj09LW7eUwB5r6Y3T28STOC2OC4UKVnLuYkJeKeoG7SkrnSsJRpLkMW7D_GuVWfw2pKEYjSNn8b9yKg_2026-02-17
Source: https://www.zdnet.com/news/rss.xml
ZDNet.com: https://www.zdnet.com
Publication date: 2026-02-17 23:51:00
Last retrieval: 2026-02-18 08:46:09.097495