Prisma AIRS by Palo Alto Networks

Genius reveals itself in the moments no one else sees coming. Just as visionaries of the past sparked revolutions with bold ideas, today’s leaders are doing the same with AI, but without the right security, can be dimmed by risk. The next era of transformation demands more than bold ideas. It demands security designed for the challenge. Prisma AIRS is built to give leaders the visibility and control to innovate fearlessly. ✔️ Discover and protect every AI app, model and dataset ✔️ Full visibility and control without slowing innovation ✔️ Security built in from the start, not bolted on later There is so much more to see at deploybravely.com ⚡ Be a Genius. Deploy Bravely.

To our incredible community at Protect AI: Thank you. This is a pivotal time in our evolution, and it’s all because of your support. As you know, we've joined forces with Palo Alto Networks. We are now thrilled to share our new name: Prisma AIRS by Palo Alto Networks. This isn't the end; it's a powerful new beginning. The journey we started together, the conversation about protecting AI at scale and by design, will only grow from here. We’ll continue this mission with the strength of the Palo Alto Networks platform behind us, bringing even greater innovations your way. We hope you'll stay with us as we continue this journey together. Let’s build the future, bravely.

🚀 See you this week at Black Hat 2025! 🚀 Join us at the AI Security Hub for expert insights and the latest in AI security innovation. What we're bringing: 🎓 Expert talks on AI security challenges ⚔️ Model Kombat battles 🏠 Hack the House interactive sessions 🐞 Between Two Vulns deep dives 🗣️ AMAs with our security experts 📍Booth #3264 📆 August 6, 10am-6pm and August 7, 10am-4pm Whether you're defending against AI-powered threats or securing AI systems, our team is ready to tackle the future of cybersecurity with you. View the full agenda: https://hubs.ly/Q03zmWHm0 #BlackHat2025 #BlackHat #AISecurityHub #CyberSecurity #AISecurity

We’re excited to be hosting Keith Hoodlet, Engineering Director of AI/ML & Application Security at Trail of Bits, for a live, AMA-style session at Black Hat USA! Join us at Booth #3264 on August 6 at 5:35 PM, where Keith will field questions on topics like: 🔹 Real-world risks in the AI development lifecycle 🔹 Securing model context and multi-agent systems 🔹 New approaches to fingerprinting AI datasets that enable teams to verify what data was used to train a model 🔹 Their journey building Buttercup, Trail of Bits’ Cyber Reasoning System competing in the finals of DARPA’s AI Cyber Challenge (AIxCC) at DEFCON 33 Bring your questions and dig into the real-world AI security challenges Keith and his team are tackling head-on. In the meantime, be sure to check out Keith’s guest appearance on the MLSecOps Community Podcast and get inspired: https://hubs.ly/Q03zQkgm0 #aisecurity #blackhat #blackhat2025 #aidevelopment #mlsecops

How do you find and fix vulnerabilities in an autonomous AI agent? Happening next week! Join Allie Howe (Founder, Growth Cyber) and Helen Oakley (Senior Director, Software & AI Security at SAP) for a live threat modeling session at booth #3264 during Black Hat USA 2025. They’ll demonstrate security risks that can emerge in multi-agent AI systems using a custom-built agent intentionally designed with known weaknesses. The session draws directly from the OWASP Agentic Security Initiative (ASI) Threats & Mitigations Guide, and highlights issues like intent manipulation, memory poisoning, and more. After walking through how these risks can surface in practice, Allie and Helen will share actionable strategies and mitigation considerations for securing agentic AI systems in real-world environments. Whether you're building agents, threat modeling LLM applications, or just beginning to explore the risks of autonomous AI, this is a unique opportunity to learn from two of the field’s most thoughtful and experienced voices. We’re deeply grateful to Allie and Helen for sharing their invaluable insights with the community! 📍 Where: Booth #3264, Black Hat USA 2025 📅 When: Wednesday, August 6 at 1:20 PM #aisecurity #agenticai #owasp #blackhat #blackhat2025

Join us and Jason Haddix, CEO of Arcanum Information Security and one of the most respected voices in offensive security, for a hands-on interactive session during Black Hat USA next week! In this deep-diving demo and learning session, Jason will walk participants through: 1️⃣ Prompt Injection Fundamentals – a guided introduction to core prompt injection techniques and attacker workflows. 2️⃣ Live AI Hacking Challenge – follow along or play independently as Jason explores a custom-built CTF experience showcasing vulnerabilities in multi-agent LLM applications. 3️⃣ Debrief & Defense – wrap up with slides and insights that connect what you’ve learned to actionable security guidance. Whether you actively participate or merely watch the action unfold, you’ll walk away with a sharper understanding of how LLMs can be manipulated & how to stay ahead of emerging threats. August 6 | 11:00 AM | Booth #3264 #aisecurity #blackhat #promptinjection #blackhat2025 #llmsecurity

🚀 See you at Black Hat 2025! 🚀 Join us at the AI Security Hub for expert insights and the latest in AI security innovation. What we're bringing: 🎓 Expert talks on AI security challenges ⚔️ Model Kombat battles 🏠 Hack the House interactive sessions 🐞 Between Two Vulns deep dives 🗣️ AMAs with our security experts 📍Booth #3264 📆 August 6, 10am-6pm and August 7, 10am-4pm Whether you're defending against AI-powered threats or securing AI systems, our team is ready to tackle the future of cybersecurity with you. View the full agenda: https://hubs.ly/Q03yMdb30 #BlackHat2025 #BlackHat #AISecurityHub #CyberSecurity #AISecurity