[REVISION HISTORY]

Balkans consumer fraud and cybercrime alerts

Updated 4 times since CLSTR started tracking revisions of this situation.

What changed

2026-08-15 16:36 UTC → 2026-08-22 11:45 UTC · added ~~removed~~

In mid-August 2026, the sophistication of digital threats in the Balkans continued to escalate. In Croatia, the Police Directorate identified a fraudulent website, ‘eprekrsaji.co/hr’, which mimics the official e-Prekršaji system to trick citizens into revealing banking information via fake traffic violation notifications. Authorities also warned of deceptive emails impersonating police officials that utilize non-official academic domains to create a false sense of urgency. Simultaneously, concerns regarding data privacy in the age of artificial intelligence have surfaced. The launch of the ‘AI Paper Trail’ tool highlighted how much personal information—including health, finances, and habits—can be inferred from digital footprints in AI chat histories. In Serbia, the Ministry of Internal Affairs issued a formal warning regarding the shift of criminal activity from physical spaces to the digital realm. Authorities noted that criminals are increasingly leveraging AI to make scams more convincing and rapid. Identified threats include the misuse of payment cards, theft of access data, fake social media profiles, extortion, and digital violence. To combat these rising risks, the Ministry encouraged the use of two-factor authentication and directed citizens to the national ‘Cyber Guard’ platform to report suspicious content or request professional assessments. By late August, new reports highlighted systemic vulnerabilities in incident detection, noting that in 31% of analyzed cases, malicious activity lasted more than three months before discovery. In Montenegro, the Cyber Security Agency issued warnings regarding sophisticated phishing scams using SMS, messaging apps, and social media. The agency noted that attackers are now utilizing artificial intelligence, voice imitations, and fake videos to make fraudulent messages appear more authentic.

Versions

Only revisions since CLSTR began indexing content versions appear here. Select a version to see what changed compared to the one before it.