Kaspersky reports 92000 AI-themed cyber attacks globally
Islamabad: Cybersecurity firm Kaspersky has reported more than 92,000 cyber attacks disguised as artificial intelligence services in 2026.
Meezan Bank crosses PKR 1bn housing finance milestone
Moreover, the findings cover the period from January to May 2026. The attacks involved malware and potentially unwanted applications targeting users worldwide.
Fake AI Apps Used as Malware Lures
According to Kaspersky, cybercriminals used trusted AI brand names to mislead users. As a result, victims were tricked into downloading malicious files.
Fake versions of ChatGPT accounted for 49% of detected attacks. Meanwhile, Claude and Gemini each represented 18% of cases.
In addition, researchers identified over 15,000 malware samples disguised as AI tools. These included spyware, banking trojans, and malware downloaders.
Advanced Threat Campaigns Identified
Kaspersky also reported a campaign linked to the Silver Fox advanced persistent threat group. The attackers distributed fake AI applications across Windows, macOS, and Linux systems.
Furthermore, the malicious software allowed long-term access to compromised devices. It also enabled theft of sensitive user information.
Expert Warning on AI Security Risks
Dmitry Galov, Head of Russia and CIS units at Kaspersky Global Research and Analysis Team, warned that AI systems are reshaping cybersecurity risks.
He explained that automated AI processes increase complexity in digital trust. Therefore, attackers are exploiting AI platforms to steal data and financial information.
He also emphasized that security must now focus on interconnected AI-driven systems rather than only endpoints.
Security Recommendations for Users and Organizations
Kaspersky advised organizations to strengthen cybersecurity using advanced protection tools. These include real-time threat detection and incident response systems.
In addition, managed security services such as MDR and threat intelligence platforms were recommended for stronger defense.
Meanwhile, individual users were urged to use only verified AI services. They should also avoid unknown bots and install trusted security software.
