Thales Unveils AI Security Fabric to Protect LLM-Powered Enterprise Applications
Thales Unveils AI Security Fabric to Protect LLM-Powered Enterprise Applications
Synopsis
- Thales introduces a dedicated security framework designed to safeguard LLM-driven and agentic AI applications
- The platform addresses emerging AI-specific threats such as prompt injection, data leakage, and model manipulation
- New capabilities focus on application security and RAG data protection, with further expansion planned
Estimated reading time: 3 mins read
Artificial intelligence is advancing at a pace rarely seen in modern enterprise technology, reshaping how organizations operate, innovate, and scale. Yet alongside its rapid adoption, AI is introducing a new class of security risks that traditional defenses were not designed to handle. Industry data cited in a report by World Business Outlook highlights that 78% of organizations now use AI in at least one business function, up sharply from 55% two years earlier. In parallel, 73% are investing in AI-specific security tools, drawing on either new or existing budgets, according to findings from the 2025 Thales Data Threat Report.
Against this backdrop, Thales has introduced the first foundational capabilities of its AI Security Fabric, a framework aimed at protecting both the core and edge of enterprise AI environments. The initiative is designed to address the growing exposure created by large language models (LLMs), generative AI, and agentic AI systems as they become embedded in critical business workflows.
The Thales AI Security Fabric is positioned as an end-to-end approach to securing LLM-powered applications, underlying data, and digital identities. Through the platform, organizations are intended to unlock AI-driven growth while keeping risks under control. This includes mitigating threats such as prompt injection attacks, unintended data leakage, manipulation of model behavior, and the exposure of sensitive or regulated information.
At the same time, the framework is built to protect data, applications, and identities across environments. Thales states that the fabric enables controlled dataset access for generative and agentic AI, applies runtime security across cloud and on-premises deployments, and safeguards AI interactions with minimal integration effort. The company also emphasizes alignment with enterprise-grade security standards, directly addressing critical risks outlined in the OWASP Top 10 and helping organizations prevent incidents that could result in financial loss or reputational damage.
The first capabilities available under the AI Security Fabric focus on two core areas. The initial component, AI Application Security, is designed to protect internally developed applications that rely on LLMs. It delivers real-time defenses against AI-specific threats including prompt injection, jailbreaking, system prompt leakage, model denial-of-service attacks, sensitive information exposure, and content moderation risks. Deployment options are intended to be flexible, supporting cloud-native, on-premises, and hybrid architectures.
The second capability, AI Retrieval-Augmented Generation (RAG) Security, addresses the data layer that feeds LLM-based applications. It enables organizations to discover and secure sensitive structured and unstructured enterprise data before ingestion into retrieval-augmented systems. This includes comprehensive data protection measures such as encryption and key management, while also securing communication between LLMs and external data sources.
Commenting on the launch, Sebastien Cano, Senior Vice President of Thales’ Cyber Security Products Business, said that as AI reshapes business operations, enterprises need security solutions purpose-built for the risks introduced by agentic and generative AI. He noted that the AI Security Fabric is intended to provide specialized protection while reducing operational complexity, drawing on Thales’ long-standing experience in cybersecurity to help organizations scale AI adoption with confidence.
Looking ahead, Thales plans to extend the AI Security Fabric in 2026 with additional runtime security capabilities. These are expected to include enhanced data leakage prevention, a Model Context Protocol (MCP) security gateway, and end-to-end runtime access control. Together, these additions are intended to strengthen protection across AI data flows, secure agentic AI access to enterprise information, and ensure unified, compliant management of interactions between users, models, and data sources.
The announcement, reported by World Business Outlook, reflects a broader shift across industries as enterprises move from experimenting with AI to operationalizing it at scale, making dedicated AI security frameworks an increasingly central part of enterprise risk management strategies.
Source here– Have a Story? Address it to the Editor and submit it here
About Thales Group
Thales is a global technology company specializing in advanced solutions across defense, aerospace, cybersecurity, digital identity, and critical infrastructure. Headquartered in France, the company operates in dozens of countries and serves governments, enterprises, and strategic industries that rely on high-assurance systems. Thales is known for combining deep engineering expertise with long-term research in areas such as secure communications, encryption, avionics, space systems, and mission-critical software.
In recent years, the company has expanded its focus on cybersecurity and data protection, addressing the growing risks associated with cloud computing, digital transformation, and artificial intelligence. Its portfolio spans identity and access management, data security, cloud protection, and now AI-specific security capabilities.
With decades of experience protecting sensitive data and national infrastructure, Thales positions itself as a trusted provider for organizations that require resilience, compliance, and security at scale, particularly in highly regulated and high-risk environments.
Featured Image: Global Morning Star
Disclaimer
The information provided in this article is for general informational purposes only and from publicly available sources. While we strive for accuracy, we do not make any representations or warranties, express or implied, regarding the completeness, reliability, or validity of the content. This article does not make any direct claims about specific companies, individuals, or organizations. Any references to reports or external sources are for context and do not imply endorsement or verification of any specific allegations. Readers are encouraged to conduct their own research and seek professional advice before making business decisions. We disclaim any liability for any losses or damages incurred as a result of reliance on the information provided.