you know, getting into the archives of the Smithsonian for genealogical records, right? You start to see after like 10 prompts in a higher incidence of the AI hallucinating, right? And what I mean by hallucinating is makes up an answer like it just gives you something and it's not necessarily true and that's where i think you know back to what daniel said it's not going to be man versus ai it's going to be people with ai and people without and the people with ai the ones that are going to thrive are going to be the ones that use it as a tool and check behind it like would any other employee

The speed in which it's changing, I mean, just while we're on this call alone, it's getting better and better.

And depending on which platform you're using, certain platforms are better at certain things.

The hallucination stuff happens, I know, on one particular platform and you've got to be careful.

There's other platforms that are better used for better industries.

And once you recognize between that and the proper way to structure your agent, to structure your bot or whatever you want to call it, you've got a prompt.

You've got to put the right prompts and guardrails in place if you're going to use it in your industry.

Like it changes every second.