In this Episode of the Practical 365 podcast, Steve and Paul deep dive into the Microsoft 365 Copilot exploit, "Echo Leak," discussing its implications for AI safeguards and the future of agentic AI. The conversation then shifts to a recent global outage affecting Google and Cloudflare, highlighting the challenges of maintaining hyperscale services. We also examine Microsoft's new initiative to offer containerized M365 instances in Europe, addressing concerns over data sovereignty. Finally, t...| Practical 365
102 posts tagged ‘prompt-injection’. Prompt Injection is a security attack against applications built on top of Large Language Models, introduced here and further described in this series of posts.| Simon Willison’s Weblog