Ask HN: How do you defend support AI agents from voice prompt injection?

3 pointstheHolyTrynity2 comments

We are securing a customer support agent that does the following: - retrieve data around company services from a RAG - is connected to a few tools to escalate to humans and write support tickets - has voice (11labs)

we did complete assessment for normal prompt injection with our tools

but is there any solution for voice attacks?

View on HN

Wait, I thought you built such a tool. 4 weeks ago you submitted "We've built an open-source tool to stress test AI agents by simulating prompt injection attacks" https://news.ycombinator.com/item?id=44060292

yes indeed, but it is not enough and was looking to find more stuff to try specifically for voice

Reply on news.ycombinator.com