Deep Instinct Enhances the Only Deep Learning-Based File Protection Solution with Launch of DPA v3.0

NEW YORK, NY, January 25, 2024 – Deep Instinct, the prevention-first cybersecurity company that stops unknown malware pre-execution with a purpose-built, AI-based deep learning (DL) framework, today announced the launch of Deep Instinct Prevention for Applications (DPA) v3.0, which includes enhanced file upload protection and application storage security capabilities, even in disconnected or air-gapped environments. DPA v3.0 is an agentless, on-demand, anti-malware solution that is device- and system-agnostic, seamlessly connecting to an organization’s existing infrastructure to quickly scan files and provide a malicious-vs-benign verdict before the file is allowed into application or storage repositories.

File transfers occur via web application uploads, user internet downloads, and third-party data sharing. Oftentimes, uploaded files aren’t being scanned or are scanned with inadequate tools like Next-Generation Antivirus (NGAV), Content Disarm and Reconstruction (CDR), or Sandbox. These solutions can’t scale to meet the volume and velocity of cyber threats brought on by adversarial AI. Additionally, these technologies have not evolved to stop unknown malware, and are largely reliant on signature-based and machine-learning (ML) models that require constant human intervention. Read the whitepaper to learn the shortcomings of these tools.

Powered by DL, DPA v3.0 is the only solution on the market that can meet these challenges and truly mitigate risks posed by file transfer. This is due to DL’s unique ability to leverage highly complex neural networks. Unlike other solutions that require near-continuous cloud calls for threat intelligence feeds and updates, DPA v3.0 only requires one to two updates per year, limiting the potential for a threat actor to capitalize on an external connection.

“With the advancements in generative AI, fueled by LLMs, adversarial AI is the biggest cybersecurity risk to organizations in 2024,” said Yariv Fishman, Chief Product Officer, Deep Instinct. “It’s imperative to fight AI with better AI, and this is where deep learning proves its superiority versus all other technologies. We’re the only solution on the market today that uses deep learning to keep data secure within both disconnected and air-gapped environments, with minimal human intervention. Businesses can freely share information and innovate with confidence, without worrying about the associated risk.”

Additional features and benefits include the following:

• Prevents unknown malware with the highest efficacy and enhanced security – With DPA v3.0, security teams get >99% efficacy in preventing unknown threats to stop ransomware, zero-day, file, and script-based attacks pre-execution.
• Enhances user experiences with near-zero latency – With a <20 milliseconds file scan time, DPA v3.0 provides a malicious vs benign verdict in the blink of an eye and eliminates the need for cloud checks or threat intelligence feeds to provide a verdict.
• Meets enterprise scale with low TCO (Total Cost of Ownership) – No matter the file size or volume, DPA v3.0 scans tens of millions of files per day while providing full privacy with only the file hash ever leaving the environment.
• Reduces the burden on the SOC (Security Operations) teams – With a <0.1% false positive rate, SOC teams can focus on real security threats vs fidelity alerts and regular maintenance requirements.
• Expands protection across more file types and sizes – DPA v3.0 supports additional file types with enhanced classification flow and better coverage of PDF, ELF, APK, MSI, DMG, and ISO. DPA v3.0 also adds support for 1TB files, 50GB archives, new deep learning models for HTML and JavaScript, and increased archive nesting levels.

To learn more about file upload protection, please download the DPA datasheet and register for our webinar on Thursday, February 8th at 11 AM ET / 8 AM PT.

For more information on Deep Instinct’s predictive prevention capabilities, visit www.deepinstinct.com.

About Deep Instinct

Deep Instinct takes a prevention-first approach to stopping ransomware and other malware using the world’s first and only purpose-built, deep learning cybersecurity framework. We predict and prevent known, unknown, and zero-day threats in <20 milliseconds, 750X faster than the fastest ransomware can encrypt. Deep Instinct has >99% zero-day accuracy and promises a <0.1% false positive rate. The Deep Instinct Predictive Prevention Platform is an essential addition to every security stack—providing complete, multi-layered protection against threats across hybrid environments. For more, visit www.deepinstinct.com.

Media Contact

Maddie Meuse
Inkhouse for Deep Instinct
deepinstinct@inkhouse.com