AI Defense API Documentation

In today's rapidly evolving technological landscape, artificial intelligence (AI) offers transformative potential for enterprises. However, the integration and deployment of AI applications introduce unique security challenges that traditional cybersecurity measures may not fully address. Recognizing this, Cisco has introduced AI Defense, an end-to-end security solution designed to empower organizations to confidently develop, deploy, and manage AI applications.

Key Features of Cisco AI Defense:

  • Shadow AI Detection and Control: Automatically identify unauthorized third-party AI applications in use within your organization, enabling enforcement of data loss prevention policies and mitigation of potential threats.
  • AI Application Security: Discover and monitor AI workloads, models, and data across distributed cloud environments. Detect misconfigurations, vulnerabilities, and adversarial attacks, ensuring AI models and applications remain secure throughout their lifecycle.
  • Real-Time Threat Mitigation: Implement guardrails to protect AI applications from evolving threats such as prompt injections, denial-of-service attacks, and data leakage, ensuring robust defense against malicious activities.
  • Comprehensive AI Asset Visibility: Gain a consolidated view of AI models, data sources, and applications across your environment, facilitating informed security decisions and effective risk management.

AI Defense API

To leverage these features AI Defense uses REST API that equips developers and administrators with powerful tools to seamlessly integrate, manage, and customize AI security protocols within their organizations.

What Can You Do with AI Defense API?

  1. Detect unauthorized AI usage – Identify and monitor shadow AI applications.
  2. Prevent data leaks – Enforce security policies to stop sensitive data exposure.
  3. Secure AI workloads – Protect AI models, data, and cloud environments from threats.
  4. Inspect AI-generated content – Analyze chat messages and HTTP traffic for risks.
  5. Detect vulnerabilities – Identify misconfigurations, adversarial attacks, and prompt injections.
  6. Monitor AI activity – Gain visibility into AI models, applications, and data sources.

See all the changes

Refer to the API Changelog for all the details.