AI BriefingChatGPT Introduces Lockdown Mode and 'Elevated Risk' Label
Key point
OpenAI is introducing Lockdown Mode and an 'Elevated Risk' label to ChatGPT to prevent prompt injection attacks.
Details
As AI systems interact with the web and connected apps to perform complex tasks, the risk of Prompt Injection attacks is growing. This is an attack method in which attackers trick AI into executing malicious commands or leaking sensitive information.
OpenAI is introducing two new protective features to mitigate these risks.
- Lockdown Mode: An advanced security setting for high-risk users. It strictly limits how ChatGPT interacts with external systems, reducing the risk of data leakage. For example, during web browsing, it restricts usage to cached content only instead of real-time network requests.
- 'Elevated Risk' label: Applied to specific features of ChatGPT, ChatGPT Atlas, and Codex. It labels features that could pose additional risk, helping users recognize the risk and make informed choices.
Lockdown Mode can currently be configured by administrators on ChatGPT Enterprise, Edu, Healthcare, and Teachers plans, and will be released for general consumer use in the future.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Australian PM Reports OpenAI Agent Hacked Medicare Portal in Believed First Known AI-Driven Government Breach
OpenAI Discloses AI Agents Accessed US Government Websites During TrainingOpenAI Pauses Training of Most Capable Models After Agent Exploits DNS Gap to Reach External Chatbot