AI-Based Phishing Increases, Cybercriminals Are Now Targeting Biometric Data

Illustration - Phishing (Photo: Freepic)

JAKARTA - Kaspersky's security solutions have detected and blocked more than 142 million click phishing links in the second quarter of 2025, an increase of 3.3 percent from the first quarter.

The global cybersecurity company revealed that the presence of artificial intelligence (AI) technology has increased phishing to a very personal threat.

AI convergences and evade tactics have turned phishing into a nearly natural, challenging clone of legal communication even for the most vigilant users, said Olga Altukhova, security expert at Kaspersky in the report.

According to Kaspersky's latest report, cybercriminals are now using deepfakes, voice cloning, and trusted platforms such as Telegram and Google Translate to disguise their attacks.

AI-generated broad language models also allow them to write messages or create seemingly convincing fake sites, without regular grammar errors being a traditional phishing signature.

Worse yet, attackers no longer only target login and passwords, but also biometric data on electronic signatures, to signatures.

Furthermore, Kaspersky also sees where the attacker is now taking advantage of the Captcha feature and Google translation service to trick security detection systems, making fake sites seem legitimate.

Using Captcha, these fake pages outwit anti-phishing algorithms, because Captcha's existence is often associated with trusted platforms, thus lowering the likelihood of detection.

Agar terlindungi dari phishing, Kaspersky menyarankan:

In order to be protected from phishing, Kaspersky suggested: