ChatGPT: In response to the growing risk of cyber threats in the digital age, OpenAI has introduced two new security features for ChatGPT: Lockdown Mode and Elevated Risk Warning System.

ChatGPT: Recognizing the growing risk of cyber threats in the digital age, OpenAI has introduced two new security features for ChatGPT: Lockdown Mode and an Elevated Risk Warning System. Both are aimed at providing users with improved security and greater transparency, allowing them to feel more in control when working with sensitive data.

What is lockdown mode and for whom is it useful?

Lockdown Mode is an advanced security option designed specifically for those at higher risk of cyberattacks. This could include executives at large organizations, cybersecurity experts, and corporate teams. This feature is not mandatory for general users but is designed for situations where an additional layer of security is needed.

When Lockdown Mode is activated, ChatGPT's interaction with external systems is significantly limited. Many tools and features are controlled or disabled to reduce the potential for data theft through hidden or malicious commands. For example, web browsing is restricted to secure cached content, and live network requests are prevented. If the security of a feature cannot be fully assured, it may also be temporarily disabled.

Additional controls at the enterprise level

Enterprise users already have robust security measures in place, but Lockdown Mode further tightens these security measures. Workspace administrators can enable it by creating a specific role. They can then determine which apps and their activities will be available in Lockdown Mode.

Additionally, compliance logging tools are provided to track which data is shared and which systems are connected to it. This helps organizations with both transparency and oversight. Currently, this feature is available in enterprise and sector-specific plans, while general consumer plans are expected to be included soon.

Elevated Risk Warning

Another major change is an elevated risk warning system. Some AI features access the internet or external systems. This increases functionality but can also pose potential risks. Some ChatGPT features will now display clear warnings to help users understand the potential impact of enabling a feature. This warning will explain the potential risks and the appropriate circumstances under which it is advisable to use that feature.

For example, if a coding tool allows network access, it will be accompanied by a warning about potential security vulnerabilities. As security technologies improve and risks are reduced, these warnings may be removed. Furthermore, new warnings may be added with new features.