Securiti revolutionizes firewall protection for GenAI systems and apps

Securiti, the pioneer of the Data+AI Command Center, is debuting the Securiti LLM Firewall, a new category of firewall designed to enhance the protection of generative AI (GenAI) systems and applications from inbound attacks. Built to understand both the context and the nature of GenAI apps at its core, the Securiti LLM Firewall represents a new frontier of protection for the AI technology sweeping every corner of industry. 

GenAI has radically transformed the nature of business, where enterprises far and wide look to adopt and implement it for greater business outcomes. However, with this advancement comes significant risk to security—especially when considering how traditional approaches to protection fail to accommodate the latest innovations in AI, according to the company. 

As Ryan O’Leary, research director, privacy and legal technology at IDC, explains, “GenAI is rapidly ushering in a new wave of innovation, but this technology also poses serious privacy and security risks that need to be addressed.” 

The Securiti LLM Firewall redefines firewall functionality, introducing innate intelligence and contextual understanding to securing conversational, multimedia, and multilingual GenAI apps. Protecting against the multitude of threats that GenAI systems are confronted with—including prompt injections, data poisoning, sensitive data disclosure, insecure output handling, jailbreak attacks, and more—Securiti LLM Firewalls detect and stop these attacks in-line and in real time across several interaction points (prompts, retrieval, and responses). 

“Securiti LLM Firewall brings unique data insights and context to the task of monitoring and controlling interactions with LLM Models, which provides more comprehensive protection for sensitive data, entitlements, and alignment with regulations in real time,” said Rehan Jalil, CEO of Securiti. 

This latest solution from Securiti embeds inspections and controls at various stages of the overall system, such as in-line next to vector databases to monitor all retrieval attempts from it. Leveraging the Securiti Data Command Graph—which provides the necessary context for enterprise controls and data—the Securiti LLM Firewall is able to protect GenAI apps with a holistic understanding of enterprise data. 

“Enterprise organizations we work with are eager to take advantage of GenAI to create business value,” said Daniel Kendzior, global data and AI security practice lead at Accenture. “Securiti’s new LLM Firewall delivers critical infrastructure to help these organizations adopt GenAI safely, mitigating privacy and security threats while accelerating innovation.” 

The Securiti LLM Firewall delivers the following advanced capabilities:

• Distributed, multi-layer protection through the LLM Firewall for prompt and response and Retrieval Firewall for data retrieved during the RAG process

• Context-aware LLM Firewalls using enterprise data to fine-tune protection, adhering to regulations surrounding sensitive data, user data rights, and compliance

• Out-of-the-box, easily customizable policies and rules that span a wide variety of OWASP GenAI threats

• Dynamic content filtering that automatically identifies, classifies, and redacts sensitive data, as well as blocks toxic content and enforces compliance with topic and tone guidelines

• A growing, comprehensive library of attacks and phishing requests to stop prompt-based attacks

• Holistic dashboard capabilities that offer transparency into enterprise AI interactions, accompanied by detailed alerts, usage insights, and policy violation tracking 

To learn more about the Securiti LLM Firewall, please visit https://securiti.ai/.