ATHENA: Preventing cyberattacks with artificial intelligence and computer vision.

Cyberattacks increasingly rely on deception to trick people into revealing confidential information or granting access to their systems. Generative artificial intelligence also makes it possible to create more convincing fraudulent messages and distribute them on a large scale. Traditional security tools may not be enough to address these threats, as they do not always identify the visual and contextual cues that can reveal a scam.

ATHENA tackles this challenge by analysing attacks from the user's perspective. To do so, it combines natural language processing, web page analysis and computer vision techniques. This approach helps detect suspicious messages, identify websites that imitate legitimate ones and assess the risk posed by each threat.

The platform combines the information gathered through these analyses to produce an overall risk assessment. It can also identify connections between different attacks and uncover potential fraud infrastructures. This helps improve threat detection and enables a faster response to potential incidents.

ATHENA provides tools to analyse suspicious web links, emails and SMS messages before they can cause harm. Its applications are particularly relevant in sectors such as banking, healthcare and industry, where protecting credentials, sensitive data and business communications is essential.