hack::soho | Agentic AI in the Wild: Risks, Reality & Framework | Max Corbridge, Secure Agentics скачать в хорошем качестве

hack::soho | Agentic AI in the Wild: Risks, Reality & Framework | Max Corbridge, Secure Agentics

Learn More: https://www.ioactive.com/ Check out this presentation, which took place at hack::soho on 30 October. This hack::soho featured a talk from guest speaker Max Corbridge, founder of Secure Agentics. The abstract of the talk, 'Agentic AI in the Wild: Risks, Reality and a Framework for Survival,' is below! hack::soho is a monthly event hosted at our London, UK office for the cybersecurity and hacking community to discuss all things security over food and refreshments. We welcome you to invite others in your circle to extend our collective network. ABSTRACT AI agents are moving fast from research labs into production, giving LLMs the power to act autonomously with tools, APIs and our data. They’re now deployed in banks, governments, and startups...but security is trailing far behind adoption. Join Max as he explains what AI agents are, how agentic AI differs from traditional GenAI, the unique attack surface they introduce, and why both adversaries and rogue behaviours are a real concern. Drawing on real-world examples, headline breaches, and current regulation, he'll introduce DAMSIC, a practical framework for secure agentic AI adoption. If you are experimenting with AI agents, or plan to in the near future, this one is for you. ABOUT THE SPEAKER Max is the founder of Secure Agentics, an AI security company helping critical organisations - including central government and tier-1 banks - adopt AI safely and securely. His recent work focuses on AI penetration testing, secure-by-design architecture, and AI security research. Before founding Secure Agentics, Max was a Principal Consultant and served as Head of Adversary Simulation leading red teams and advanced penetration testing functions. He has delivered complex security engagements across cloud, application and infrastructure domains, advised boards on cyber risk, and published research that uncovered high-impact vulnerabilities reported worldwide.