Artificial intelligence...a hidden threat to cybersecurity in Egypt

Listen to the article. The audio text is automated, generated by an automated system

The most dangerous cybersecurity threats no longer necessarily come from across borders or through breaches of digital firewalls. Often, an open browser window may suffice for an employee seeking to complete his tasks quickly, using a free artificial intelligence tool that the organization does not even know exists.

This hidden threat represents a new and more complex facet of data security risks in the contemporary business environment.

Dr. Mohamed Mohsen Ramadan, Head of the Artificial Intelligence and Cybersecurity Unit at the Arab Center for Research and Studies, confirms in exclusive statements to “Al Arabiya.net/Al Hadath.net” that this phenomenon has evolved from a traditional concept previously known to technology, represented by the use of unapproved software, to a more dangerous level, given the ability of artificial intelligence systems to process data and analyze texts, databases, and code codes.

Ramadan explained that the irony is that the employee may have good intentions, and only aims to reduce hours of work to minutes, by summarizing a report, correcting a software code, or analyzing customer data. But he may be completely unaware that the problem is not only related to the result he obtains, but rather extends to the exit of sensitive data from the institutional environment to unsafe external services.

According to Ramadan, recent data indicates the size of the gap between the speed of technology adoption and the absence of a governance system, as the use of generative artificial intelligence applications within institutions increased from 74% in 2025 to 96% in 2026.

In the same context, 38% of employees reported that they shared sensitive information related to working with artificial intelligence tools without their employers’ permission, according to what the Egyptian specialist indicates.

The most dangerous scenario of unintentional data leakage emerges from this, such as human resources departments uploading CVs, or legal departments reviewing contracts via external platforms.

For his part, Major General Mohamed Rajai, former Assistant Minister of the Interior of Egypt, explained to Al Arabiya.net / Al Hadath.net that cybersecurity is no longer limited to preventing the attacker from entering, but also extends to preventing sensitive data from leaving in an unintended manner.

He added that the seriousness of these incidents lies in the fact that they appear to be completely normal activities for the user: entering a site, pasting text, getting an answer, then closing the page.

This makes traditional protection tools unable to detect them, in the absence of clear indicators of penetration, so that the well-intentioned employee becomes an invisible leak channel.

The Egyptian security official warned against dealing with artificial intelligence with the logic of complete prohibition, explaining that blindly prohibiting the use of applications often pushes the use to greater secrecy and concealment.

He said that instead, we must move to a philosophy of safe use and smart governance, by building a clear map of the uses of artificial intelligence, classifying tools and data within the organization, and precisely defining permitted and prohibited tools, while providing safe and approved alternatives to employees.

Major General Rajai also called for controlling the powers of smart systems and agents, to ensure that they do not access sensitive files without oversight, in addition to incorporating training and cyber awareness as a basic line of defense to establish the limits of safe use among workers.