Corey Knowles

Tell them the show sent you.

It did not do well in the world of safety, did it?

Yeah, at least you know what you're getting into then.

Like you're able to learn more about the model.

You're able to see it's testing info and model cards, I assume.

I guess that's probably a big difference maker.

You know, I noticed on Bloomberg, you called the AI action plan a moonshot moment.

For our listeners who haven't read the full document, what's the big picture here for you?

That does.

Would you think of that in terms of like just for some context this morning, OpenAI dropped a pair of open source models not long before we're recording this.

And I realize that there's minimal context at this moment given the amount of time.

But is maybe the answer open source coming from companies that are beholden to some governance in themselves versus...

just out on the open market because, you know, the one thing that concerns me is I am pro-open source for all of the reasons that you should be pro-open source.

But at the same time, I am equally cognizant of the fact that once it's out there, you don't pull it back, you know.

So let's think of this in practical terms then.

Suppose I'm running a small business and I want to use AI for something simple like customer service emails versus processing insurance claims.

Those need very different types and levels of oversight, correct?

What does that different look like?

I'm assuming we're talking about various types of compliance that already exist as well as something new?

That makes a lot of sense.