So, how do I know if SuperSim is right? And what happens when I get something like this in ChatGPT, which is what if I get something wrong?

Sure.

I think there can be a whole stand-up routine on the way AI goes wrong and the way it hallucinates and the way it tortures us with its answers.

Um, but jokes aside, I think SuperSim, the way we've built it, it's a combination, of course, of an LLM, so we are using frontier LLM capabilities, including OpenAI.

So, you get that power inside SuperSim.

But in addition to that, it's trying to capture tribal knowledge, so knowledge that isn't really written down in the documents.

It actually tries to identify what might be missing and, and kind of elicit it through, through its conversator- conversations.

So what do you think of Taiwan's new AI law about transparency, explainability, and keeping humans in control? Do you think Taiwan is thinking of these problems right now?