Noah Labhart

And I'll just jump right into my first question.

What are inference time threats?

That'd be the best place to start is to define what those are.

And why are they becoming a critical focus in AI security?

That makes sense.

And where you're trying to protect it makes sense.

And I think it's important to maybe tell the difference between inference time risk and training time risk as well.

Can you differentiate the two?

I think my next question, you may have touched this at a high level, but I want to give it space here.

Why is protection from these in-the-moment risks, right, with inference time, why is protection from those critical for safe, scalable AI adoption?

And as an engineer myself, I think that I could come up with some ideas of what you're going to say, but I'm curious what you have for the audience.

Interesting.

Okay, that's super interesting.

I think to move into the next area, right, of my questioning is more around industry, right?

And perhaps these threats are different across industries.

Maybe they're not.

I guess that's the question.

Do they vary across industries?

And if so, is there an industry where these attacks are more prevalent?

That makes a ton of sense.