AI Confidence: It’s a Trap! (LIVE in San Francisco)
Early research is showing that Ai provides clear productivity gains, but the ROI is harder to pinpoint. Almost all LLMs will confidently produce seemingly high-quality output. But are we setting ourselves up for failure when we mistake confidence for accuracy?
This episode was recorded in front of a live audience at BSidesSF 2026.
Click here to see more photos from our live podcast recording and CISO Series game show at BSidesSF 2026.
Join the conversation on LinkedIn.
Huge thanks to our episode sponsors, Nudge Security, QuilrAI, and Zenity
Get a full inventory of AI assets on Day One of your free trial, even those introduced before you started using Nudge. Get started at https://www.nudgesecurity.com/
Can you tell if an action in your environment was performed by a human — or an AI agent? QuilrAI’s Decision Engine evaluates content, context, and intent before actions complete — across browsers, endpoints, SaaS, LLMs, and agents. Not more alerts. Better decisions, in real time. Visit quilr.ai.
Help shape the future of AI agent security. On May 27th, the AI Agent Security Summit returns to San Francisco. Hear from leading researchers and security pioneers, and usher in the new age of secure AI deployment across the enterprise. Register at Zenityzenity.io/ai-security-summit.