Creating a lock in ChatGPT: What is it, and how does it help to protect sensitive data?
We are increasingly relying on artificial intelligence in many areas, making it crucial to ensure the security of every action. In this context, a threat known as "instant injection" has emerged. This type of attack involves attempting to trick the system into executing malicious instructions or revealing confidential data.
Therefore, to enhance protection against these threats, OpenAI has launched new security measures specifically designed for environments where data protection is a top priority. This option does not change how ChatGPT is used, but it restricts some functions to reduce the risk of leaks or external manipulation.
How to enable lock mode in ChatGPT
According to Sam Altman, ChatGPT's key new feature is Lock Mode, an optional advanced security setting designed for users who handle sensitive information, such as business owners, executives, security teams, and organizations in the education and healthcare sectors. While not essential for everyday chatbot users, it can be useful in the event of a potential malfunction with serious consequences.
They explain that locking mode limits ChatGPT's ability to interact with the outside world, reducing or disabling certain functions that could be exploited in an attack. For example, when browsing the internet in this mode, the system only accesses pre-cached content and does not make real-time connections outside the OpenAI environment. Furthermore, some tools are completely disabled if data security cannot be fully guaranteed.
How to access lock mode in ChatGPT
This mode is available in business and professional packages such as ChatGPT Enterprise, ChatGPT Edu, Healthcare, and ChatGPT for Teachers. To activate it, the administrator must log in to the workspace, create a new role, and enable lock mode.
He can then specify which applications and actions are allowed. This gives the organization simple and precise control over how the tool is used and how data is protected.