How ChatGPT's New Lockdown Mode Protects You From Data Theft (and What Else It Does)
The mode disables browsing, agents and file downloads for logged-in users, while OpenAI says it reduces but does not eliminate exfiltration risk.
- OpenAI began rolling out Lockdown Mode in ChatGPT to protect users from prompt injection attacks, disabling live browsing, agent mode, and deep research across Free, Go, Plus, Pro, and self-serve Business plans.
- Security researchers have demonstrated prompt injection hijacks against agents from Anthropic, Google, and Microsoft via GitHub Actions integrations, as attackers hide malicious instructions in webpages and files to steal sensitive data.
- Specifically, Lockdown Mode blocks live web access, network-connected Canvas functions, image retrieval, and file downloads; by cutting off outbound network requests and pixel-based data channels, the mode substantially reduces attack surface.
- OpenAI cautions that Lockdown Mode is not a complete defence against prompt injection attacks, as malicious payloads in cached webpages or uploaded PDFs can still influence model behaviour, though the feature blocks outbound data channels.
- The underlying weakness remains fundamental: LLMs cannot reliably separate data from instructions. For sensitive-data handlers the trade-off is worth making, but the expanding agent ecosystem means risk continues rising for other users.
27 Articles
27 Articles
OpenAI announced the expansion of one of its most advanced security features to ChatGPT. The company began implementing Blocking Mode on all personal accounts of the service, including...
OpenAI adds Lockdown Mode to ChatGPT to block data theft from prompt injection attacks
OpenAI has begun rolling out Lockdown Mode to ChatGPT, a new security setting designed to block attackers from stealing data through prompt injection attacks. The feature disables live web browsing, agent mode, deep research, image retrieval, Canvas networking, and file downloads. It is available to logged-in users across Free, Go, Plus, Pro, and self-serve ChatGPT […] This story continues at The Next Web
OpenAI unveils lockdown mode feature to protect sensitive data from prompt injection attacks
The feature, which is available to users on business plans like Free, Go, Plus, Pro and self-serve ChatGPT, is designed for individuals and organisations desiring stricter controls of how sensitive data can be accessed and transmitted through OpenAI products.
OpenAI introduces an optional lock mode for ChatGPT to specifically minimize the risk of data outflows due to prompt injection attacks. Software developer OpenAI has begun the gradual introduction of a new security feature called Lockdown Mode (block mode) for ChatGPT. This optional, extended setting is available to logged-in users of various subscription models, including Free, Go, Plus, Pro and the self-managed ChatGPT business plans. The feat…
Coverage Details
Bias Distribution
- 60% of the sources lean Left
Factuality
To view factuality data please Upgrade to Premium
