The Ledger
Every statement that passed quotation and attribution checks. Mix any filter with any other: certainty 1/5, debate potential 5/5, or both at once.
why aren't all 7 resolved? a statement only gets an assessment when the public record can support or contradict it. opinions and what-ifs never can, and 0 checkable ones are still open, waiting for their date. predictions held up or didn't; assertions are supported or contradicted. on every card: ▮▮▮▮▮ certainty · ▮▮▮▮▮ debate potential. speakers are clickable
Emmons: Prompt guardrails fail to reliably constrain AI agents
“So stepping in there with policy that says do not go on this, that's a big deal for things in the medical setting where you may have, you know, people use the word guardrails to talk about AI where they're like, we're going to put guardrails so it doesn't do t…”
Emmons: Requiring humans to review every AI output eliminates ROI
“If a whole, if a human's now tasked with reviewing everything, we find a lot of times say you're doing a claims adjustment or underwriting an insurance place capacity, you would have where you're not making the value back because the person that's checking it …”
Emmons: AI agent ROI requires deep understanding of internal workflows
“The companies who are winning are the ones who have excellent understanding of their internal operations of how these things work, and they can go, well, we do this, and it's not quite an assembly line. There's still creativity and maybe the results or the jud…”
Emmons: Developers should start with giant AI models before pruning to niche ones
“So my suggestion usually to everybody is start with the biggest and you can test the smaller ones to see if they keep up or keep well enough for what you're trying to do. But if it really matters, start with the biggest and then use a tool like swept to speed …”
Emmons: Test AI at the scoped agent level, not foundation models
“When we're testing them and trying to figure out how they function, one of the first things we do is we don't go down to the model level, the chat GPDs, the clods, the Geminis of the world, because they have You can think of it as, like, infinite possibility. …”
Emmons: Swept was founded after an AI agent harmed teenagers
“I mean, we originally were started because we saw an AI agent perform some pretty bad harm to some teenagers that were trying to seek help. And that's where we were like, well this is going to be a common pattern. So let's make sure it doesn't go that way.”
Emmons: Swept wasted months chasing VC over business fundamentals
“We started out bootstrapping, and then we got it in our minds that we're a startup. We need to go get raised venture capital, and I think we spent Three to four months too focused on trying to raise venture capital and not enough focus on building good fundame…”