Learn from Microsoft’s AI Red Team on how to make your organization safer - Seguridad Mania.com - España y América Latina

Microsoft believes in developing responsible AI, and that belief extends to all partners across the industry. But with 3.5 million unfilled jobs in the cybersecurity industry, not every company has the resources to develop their own specialized teams to red team AI and LLM products. That’s why Microsoft’s industry-leading AI Red Team aims to share its learnings and best practices with the cybersecurity community. With this in mind, Microsoft is releasing PyRIT (Python Risk Identification Toolkit for Generative AI), an industry-first automation framework developed by the Microsoft AI Red Team aimed to help enhance the safety and security of LLM endpoints. PyRIT from the Microsoft AI Red Team is an open-source toolkit to empower security professionals and machine learning engineers to proactively find risks in their generative AI systems. PyRIT is battle-tested by the Microsoft AI Red Team in over 60 red teaming exercises of generative AI systems. It is meant to supplement manual probing and help identify risks in generative AI systems faster. It can help change its tactics based on the response from the generative AI system and generate the next input to the generative AI system. This automation continues until the security professional’s intended goal is achieved. In this webinar, we will walk through how to use PyRIT to red team a generative AI system – how to set up targets, how to leverage the datasets, different attack strategies, and the memory functionality. Come learn from Microsoft's best practices on how you can empower your Red Teams and make your organization safer.