Published • loading... • Updated
OpenAI rolls out a Lockdown Mode for extra protection against prompt injection attacks
The optional setting limits network requests and disables some tools, and OpenAI says it is aimed at users handling sensitive data.
- OpenAI has rolled out Lockdown Mode, an optional security setting designed to offer users advanced protection against prompt injection attacks.
- As AI usage has skyrocketed in recent years, prompt injection attacks have become a critical security issue where attackers hide malicious instructions within content like PDF files or websites.
- Enabling this mode limits outbound network requests and disables features like Deep Research and Agent Mode, restricting web browsing to cached content only.
- While the feature reduces the likelihood of attackers extracting sensitive data, OpenAI notes it does not affect conversation memory or file uploads; workspace administrators retain separate controls in enterprise settings.
- Designed specifically for handling sensitive data, this tool is not intended for everyone, as most users do not need it; users can enable it via ChatGPT Settings under Advanced Security.
Insights by Ground AI
14 Articles
14 Articles
New ChatGPT Lockdown Mode to Mitigate Prompt Injection and Data Exfiltration Attacks
OpenAI has released ChatGPT Lockdown Mode, a new security feature designed to limit outbound network access and reduce the risk of data exfiltration from prompt-injection attacks. The feature is now available to eligible personal accounts, self-serve ChatGPT Business users, and managed enterprise workspaces. Prompt injection, where malicious instructions are embedded in content processed by an […] The post New ChatGPT Lockdown Mode to Mitigate P…
All ChatGPT users now have access to the so-called lockdown mode. The function improves security by protecting them against so-called prompt injection attacks. (Continue reading)
Coverage Details
Total News Sources14
Leaning Left3Leaning Right1Center0Last UpdatedBias Distribution75% Left
Bias Distribution
- 75% of the sources lean Left
75% Left
L 75%
R 25%
Factuality
To view factuality data please Upgrade to Premium
