To help make AI safer for everyone, we’ve shared the SAIF Risk Assessment, SAIF Risk Map, and the associated risks, controls, and components descriptions with the Coalition for Secure AI (CoSAI). We hope this contribution accelerates the development of open-source security solutions, making it easier for any organization to build more secure AI systems.
SAIF Map
The SAIF Map is a visual guide for navigating AI security and is at the heart of understanding SAIF as a security framework.
Many AI risks are new to developers, such as Prompt Injection, Data Poisoning, or Rogue Actions. The SAIF Map shows the AI development process to locate the risks—and crucially, potential controls for those risks.
How to use the SAIF Risk Map
You can use the SAIF map to:
- follow a single risk throughout the AI development process
- see the specific risks that affect a certain component area or type of organization
- or better understand how your own organization’s development structure might be affected by AI risks
The SAIF map is divided into four component areas: Data, Infrastructure, Model, and Application. Some approaches to AI security focus primarily on the model, whereas SAIF addresses risks and controls throughout the entire AI development lifecycle. For advanced systems like AI agents, SAIF also offers a second, specialized diagram that goes into greater detail on the unique risks they introduce.
The top half of the map shows the path a model takes to deployment in an application and how a user queries the model through that application. This content is most relevant to Model Consumers—those who use AI models to build AI-powered products and applications.
The bottom half of the map shows the path to developing a model and is most relevant to Model Creators—those who train or finetune models for use by themselves or others.
Depending on how you use AI, certain risks may be more relevant to you than others. Use the SAIF Map and the Risk Self Assessment to discover which risks you should investigate.