The government has introduced stricter rules for officials using artificial intelligence tools, particularly when handling sensitive information.
Under the new guidelines, employees have been barred from uploading classified documents to public AI platforms. The move aims to reduce the risk of confidential information being exposed through AI tools.
National CERT Issues New AI Security Guidelines
The National CERT has issued the National Cyber Security Handbook 2026-27, which incorporates the principles of the Pakistan Information Security Framework 2026.
The handbook sets out restrictions on how government employees should use public AI platforms. In particular, officials cannot upload official emails or software source code to such tools.
The restrictions also cover citizens’ personal information. Government employees have been instructed not to upload personal data to public AI platforms.
According to the National CERT, sensitive information should not be entered into AI systems because it could lead to data leakage. Therefore, officials have been directed to use only AI tools approved by their respective departments.
Officials Told to Remove Sensitive Details
The new guidelines also require officials to exercise greater caution when preparing AI prompts and uploading files.
Officials have been instructed to remove names and other sensitive details before submitting information to AI platforms. This requirement applies to both prompts and files used with AI tools.
Furthermore, any accidental disclosure of confidential information must be reported immediately to the relevant cybersecurity team.
The guidelines also prohibit government employees from sharing passwords, administrative login credentials or API keys with AI tools. These restrictions are intended to prevent sensitive access information from being exposed through AI platforms.
Unapproved AI Extensions Also Banned
The handbook also addresses the use of AI-related software on official devices. Employees have been prohibited from installing unapproved AI extensions and plug-ins.
In addition, any AI-generated content used for government work must undergo checks for accuracy and security. Officials cannot rely on AI-generated material without reviewing its reliability and potential security risks.
The National CERT has further stressed that AI-generated content cannot be used in official matters without human oversight.
Therefore, the use of AI within government departments will remain subject to security, privacy and supervision requirements.
Overall, the new guidelines place greater responsibility on government employees when using AI tools. Officials must ensure that sensitive information stays protected while any AI-generated material receives appropriate human review.
For the latest updates, visit and follow The Truth International website (www.thetruthinternational.com) and subscribe to the YouTube Channel.
