Kara Santamaria

So more actors may, may engage in,

Now, this has been blown wide open with LLMs entering the picture because now a naive user, and there are studies that are beginning to compile to show this, a naive user can produce novel LLMs.

and let's just say dangerous on a lot of levels, different pathogens.

And specifically, there's a lot of research now into the development of proteins.

Proteins before were a huge bottleneck because they're complicated.

Proteins are one of the most complex macromolecules in nature, and they do so much.

Proteins are just, they're ubiquitous.

They're involved in so much metabolic activity.

Everything, so much, they're all there.

If you say protein, you're probably going to be right.

Yeah, exactly.

Most every enzyme is a protein.

Proteins do a lot of stuff.

So I did a little bit more digging, and I found a really great kind of position paper from an individual ethicist who writes about technology and the risks of powerful AI.

And he basically detailed some of the benefits and some of the concerns and some of the things that we should be thinking about and highlighting.

So these are the things he said.

When we think about the benefits, it's important to discuss risk in a careful and well-considered manner.

I think it's critical to, number one, avoid damage.

Not just in the sense of believing doom is inevitable, but also more generally thinking about AI risks in a more religious way.

He said we should acknowledge uncertainty.