Cybersecurity company Barracuda Networks Inc. today unveiled new threat detection capabilities fueled by multimodal artificial intelligence that deliver context-aware protection against emerging attacks.

The new capabilities give Barracuda’s platform the ability to protect against attacks with accuracy and speed by simultaneously correlating and analyzing diverse text and visual data types – including URLs, documents, images, QR codes and more.

Built on Barracuda’s existing use of AI and machine learning, the new capabilities introduce a new integration with multimodal AI, technology that synthesizes and interprets numerous data streams in various formats, with machine learning classifiers and a purpose-built sandbox engine. Doing so delivers a faster, smarter and more adaptive defense layer that detects more than three times as many malicious files at eight times the speed of previous models, the company says.

“As cybercriminals increasingly weaponize AI to launch targeted, sophisticated and evasive attacks, businesses need security that’s equally intelligent and adaptive,” explains Sunil Kumar, vice president of advanced technology at Barracuda. “By leveraging multimodal AI, we’re taking a holistic approach to analyze different data types to identify attacks that bypass traditional models. It’s a transformative step forward in proactive cybersecurity.”

The capabilities significantly strengthen Barracuda Advanced Threat Protection, which provides layered security across the Barracuda platform. The new capabilities also enhance Barracuda LinkProtect, which inspects URLs for hidden threats, malicious scripts, suspicious redirects and other attacks using a virtual sandbox and secure, isolated browser environment. Real-time threat intelligence is continuously shared across the entire Barracuda platform to deliver automated, adaptive defense at every layer.

The rise of generative AI and deepfake technologies has added complexity to the threat landscape, making traditional security models increasingly insufficient on their own. By incorporating multimodal AI, Barracuda can now detect not only known threat signatures but also subtle anomalies across multiple content formats that may indicate novel or zero-day attacks. The capability is particularly useful at a time when threat actors continue to blend social engineering, image-based lures and malicious links into more convincing and evasive campaigns.

Image: SiliconANGLE/Reve