> For the complete documentation index, see [llms.txt](https://cortex-docs.paloaltonetworks.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://cortex-docs.paloaltonetworks.com/cortex-cloud-posture-management/cortex-cloud-ai-security/cortex-cloud-ai-security-use-cases.md). # Cortex Cloud AI Security use cases #### Understand your AI ecosystem Understanding your AI ecosystem is crucial for identifying potential vulnerabilities and ensuring the robustness of your AI operations. A comprehensive view of your AI landscape helps in pinpointing where sensitive data is processed and stored, as well as how data flows between systems. To understand your AI ecosystem, use the **AI Security Dashboard**, which provides visibility into all the AI components. You can also see how your AI assets relate to any other asset in the environment using the **Graph Search**. The complete list of your AI assets can be found under **AI Inventory**, where you can investigate each asset. #### Investigate an AI asset To understand a specific component of your AI ecosystem and identify any findings or security issues related to it, use its asset card and links to findings, issues, and cases created for the asset. When you select an asset, you can review all the tabs on its asset card. These tabs on the asset cards provide information about the following: overview, access, data, vulnerabilities, applications, and AI ecosystems. #### Detect AI security issues Detecting the AI security issues early is pivotal to safeguarding AI-powered applications and the sensitive data they handle. AI systems, due to their complexity, can often be opaque, making it difficult to identify vulnerabilities using traditional methods. To detect security issues in your AI ecosystem, use the **AI Security Issues** page. #### Secure the data for AI Securing the data utilized by AI systems is critical. Cortex Cloud AI Security helps you identify the data that is impacted by your AI ecosystem, whether it's training data, data used for RAG (Retrieval Augmented Generation) or any other related data such as prompt logs. It also classifies this data, using Cortex Cloud Data Security. Data classification across your AI ecosystem allows you to identify models that are trained on sensitive data and to prioritize all identified risks and issues based on their data impact. For example, missing guardrails on a sensitive model should be treated differently due to its context. #### Discover self-managed AI models Cortex Cloud AI Security helps organizations discover self-managed AI models. Self-managed AI models refer to AI models that are deployed and operated on self-managed cloud infrastructure, rather than through cloud providers' managed services. These models are often sourced from public repositories like Hugging Face, and can lead to the proliferation of shadow AI. The growing use of AI in business workflows makes it increasingly important to manage and secure all AI models, whether deployed through managed services or self-managed infrastructures. Self-managed AI models, in particular, introduce unique risks, such as security vulnerabilities and compliance gaps. Tracking and securing these models is essential to reducing risks and ensuring that AI applications remain safe, secure, and compliant. #### Comply with AI regulations Cortex Cloud AI Security ensures compliance with emerging AI mandates and industry standards, which is crucial because new frameworks require unique measures to govern AI-specific vulnerabilities. For example, data poisoning is a major risk for AI applications but traditional compliance programs are not designed to handle it; however, new frameworks for AI governance include relevant measures, such as the documentation of data sources used to train AI models. In addition, AI-powered applications also add complexity for existing regulations like GDPR, due to their data processing and interconnected systems. Cortex Cloud AI Security allows for continuous monitoring and visualization of compliance with leading AI standards, such as the OWASP Top Ten for LLM. Complying with current industry standards can help shorten the time needed to meet future binding regulations. Cortex Cloud AI Security helps you enforce policies, maintain audit trails, and achieve compliance, providing visibility into compliance violations and helping manage your AI Inventory, which is essential for controlling model sprawl and shadow AI. #### Manage your AI software supply chain As AI becomes deeply embedded in application development, security teams need comprehensive visibility into the software supply chain. This visibility must go beyond deployed AI models and agents, extending to the underlying AI software packages and SDKs that developers use to build these systems. A key aspect of Cortex Cloud AI Security is implementing a shift-left approach to AI security. This helps organizations identify and manage risks early in the development lifecycle by providing visibility into the AI software supply chain. Understanding this supply chain is crucial for both generating an AI Bill of Materials (AI-BOM) and for identifying potential vulnerabilities before they are deployed to production. This proactive stance ensures that security is addressed at the source, preventing more complex and costly issues later on. #### Detect open-source models Cortex Cloud AI Security provides detection and risk assessment for open-source models, identifying and displaying the count of open-source models on the dashboard. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://cortex-docs.paloaltonetworks.com/cortex-cloud-posture-management/cortex-cloud-ai-security/cortex-cloud-ai-security-use-cases.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.