Good Vibes Only: Security with AI Vibe Coding

As large language models (LLMs) and AI tools become increasingly powerful in the development world, a new concept has emerged: vibe coding, or guiding an AI to build out code without ever laying a finger on the code yourself.

But as with most shiny new tools, security is often left behind in the rush to innovate. AI-generated code can be incredibly helpful, but it rarely defaults to being secure unless explicitly guided. Beyond the usual concerns of bugs and bad actors, vibe coding also introduces risks around copyright, attribution, API usage limits, and data privacy.

This session explores the hidden (and not-so-hidden) security pitfalls of AI-assisted development. You'll walk away with a mild amount of paranoia, practical tips, go-to resources, and solid practices to help ensure that your next side project, startup prototype, or enterprise application is built on secure foundations.

If you're going to let an AI build code, at least make sure it’s not building your next security incident as well. Let the good vibes roll!

• Development
• Other