Apiiro introduced Risk Detection at Design Phase, a new, AI-driven capability that automatically analyzes feature requests to identify risks and proactively initiate security reviews or threat models at the earliest stage of the application development lifecycle. With this new, first-of-its-kind capability, application security (AppSec) practitioners can now scale their secure software development lifecycle (SSDLC) processes by mitigating security and compliance risks before a single line of code is written. 

Security products on the market today detect risks only after the development process has begun. This results in wasted time for developers due to manual risk assessment questionnaires, which impact release velocity and business value. With the detection of risks at the design phase, Apiiro customers can proactively address security, data privacy, infrastructure, compliance, and other risks at the onset of development, saving significant time and costs while minimizing rework and accelerating secure software delivery. 

Apiiro's detection of risky feature requests is built on cutting-edge AI technology, including Apiiro's native private LLM. This model, not accessible by ChatGPT or any other public LLM services, ensures customer privacy and compliance by automatically analyzing feature requests and proactively identifying potential risks associated with:

• Architecture design and security controls: requests for changes in APIs, network, databases, web servers, web clients, logging, serialization and other component configurations, architecture designs, and deployment of new or changed components. 
• Sensitive data handling: storing and/or processing sensitive information like PII, PHI payment data fields as part of the application data flow, changing encryption mechanisms, data migrations, writing sensitive data to logs, and using sensitive data as an API return type.
• User permissions and access management: user authentication and authorization, login or registration processes, and changing user permissions. 
• Generative AI technology: adding or changing generative AI tools, frameworks, technologies, and the data that is exposed to them.
• Third-party integrations, and open source dependencies: changing or adding open source dependencies and integrations with third-party services.

For each risky feature request, enriched by the code architecture generated by its Deep Code Analysis (DCA) technology, Apiiro's native private LLM model automatically generates contextual questions for a security review and produces threat stories using the STRIDE model. This automation eliminates the need for manual security processes, accelerating development velocity and deployment of secure code to the cloud, ultimately driving business growth. In addition, Apiiro enhances design risk context by automatically mapping to specific code commits, repositories, and pull requests, providing deeper insight into how potential risks may manifest in the actual codebase. 

"Detecting potential risk at the design phase gives us the opportunity to remediate risks before they exist, and in the most efficient way for our developers. However, it's challenging to do this at scale and to ensure full coverage of features our development team are building. Apiiro's design phase risk detection engine is a unique capability in the ASPM space. It allows us to modernize our approach to Secure-by-Design, scale and strengthen our security engagement, and provide some automation to our threat modeling and security requirements processes." -Head of Security Engineering at Fortune 100 retail company

"Amidst the ever-changing complexity of modern software development processes and application architectures, Apiiro is committed to delivering complete risk-based visibility and protection from design to runtime," said Moti Gindi, chief product officer at Apiiro. "Building secure software starts with secure design, and the new AI-Driven Risk Detection at Design Phase from Apiiro takes the ‘shift left' approach a step further, addressing risks even before a single line of code is written. This first-of-its-kind functionality leverages the power of AI to ensure customers have the context required to facilitate efficient security reviews and evolve from a reactive to a proactive approach to application security."