Cisco Unveils AI Defence to Secure Enterprises’ AI Transformation

• Cisco introduces an end-to-end solution designed to protect the development and usage of AI applications, enabling enterprises to advance their AI initiatives with confidence.
• AI Defence mitigates risks such as AI tool misuse, data leaks, and sophisticated threats that traditional security systems cannot handle.
• The solution leverages Cisco’s extensive network visibility and control, addressing evolving concerns around AI safety and security.

Singapore, 21 January 2025 – Cisco (NASDAQ: CSCO), a global leader in security and networking, has announced the launch of Cisco AI Defence, a cutting-edge solution designed to support and safeguard the AI transformation journey of enterprises. With the rapid advancement of AI technologies, businesses face unprecedented safety and security challenges that existing solutions are not equipped to manage. Cisco AI Defence aims to empower enterprises to confidently develop, deploy, and secure AI applications.

Jeetu Patel, Executive Vice President and Chief Product Officer at Cisco, commented, “Business and technology leaders can’t afford to sacrifice safety for speed when embracing AI. In a dynamic landscape where competition is fierce, speed decides the winners. Fused into the fabric of the network, Cisco AI Defence combines the unique ability to detect and protect against threats when developing and accessing AI applications without trade-offs.”

The Rising Stakes of AI Security

The potential repercussions of AI-related security breaches are significant. Cisco’s 2024 AI Readiness Index reveals that only 28% of respondents in Singapore feel adequately prepared to detect and prevent unauthorised tampering with AI. The complexity of AI applications, often operating across multiple models and cloud platforms, introduces vulnerabilities at the model or application level. With responsibilities spread across developers, users, and vendors, securing proprietary data becomes increasingly critical as enterprises move beyond public datasets.

To accelerate AI innovation and adoption, businesses require a unified layer of safety and security that safeguards all users and applications. Cisco AI Defence addresses two critical risks:

1. Developing and Deploying Secure AI Applications

As AI integration becomes widespread, enterprises are expected to manage hundreds or even thousands of AI applications. Developers need a unified set of security guardrails that function across platforms. Cisco AI Defence offers:

• AI Discovery: Helps security teams identify who is building applications and the sources used for training data. It detects both authorised and unauthorised AI applications across public and private clouds.
• Model Validation: Automated testing ensures AI models are free from toxic or unexpected behaviours by checking for hundreds of safety and security issues. This process leverages an AI-driven algorithmic “red team” to identify vulnerabilities and propose safeguards.
• Runtime Security: Provides continuous protection against threats such as prompt injection, denial-of-service attacks, and sensitive data leaks.

2. Securing Access to AI Applications

As users increasingly adopt AI tools for productivity, security teams must prevent data breaches and the contamination of proprietary data. AI Defence delivers:

• Visibility: A comprehensive overview of both sanctioned and unsanctioned AI-enabled applications used within an organisation.
• Access Control: Policies to regulate employee access to unauthorised AI tools.
• Data and Threat Protection: Continuous monitoring to prevent data loss and ensure compliance while defending against evolving threats.

Unmatched Expertise and Integration

Unlike guardrails built into individual AI models, Cisco provides consistent controls for a multi-model environment. AI Defence self-optimises using Cisco’s proprietary machine learning models to address emerging AI safety concerns. By integrating threat intelligence from Cisco Talos, the solution offers enriched alerts and seamless visibility across the entire ecosystem. Designed within Cisco’s Security Cloud, an AI-driven cross-domain security platform, it utilises Cisco’s vast network of enforcement points to deliver unparalleled AI security at the network level.

Cisco has also played an active role in shaping AI security industry standards through collaboration with MITRE, OWASP, and NIST. The solution underscores Cisco’s commitment to ensuring accuracy and trustworthiness in protecting enterprise AI applications.

Kent Noyes, Global Head of AI & Cyber Innovation at World Wide Technology, remarked, “The adoption of AI exposes companies to new risks that traditional cybersecurity solutions don’t address. Cisco AI Defence represents a significant leap forward in AI security, providing full visibility of an enterprise’s AI assets and protection against evolving threats.”

A Step Forward in AI Security

Cisco AI Defence is the latest in a series of innovations, including Cisco Hypershield, aimed at advancing AI-driven security solutions. It will be available to enterprises starting March 2025. For further details, visit cisco.com/go/ai-defense.