The Exchanges

Every argument clarity score on this site is built from rows on this page. Each question and answer was assessed with names hidden, the host's own answers included, on four things from 1 to 5: directness (does it answer the question asked), coherence (do the ideas follow), precision (concrete details and clear references), compression (says a lot per word). The weighted mix (30/30/25/15) is the exchange score. A person's published score averages their exchange scores on raw tape only, at least 8 of them, shrunk toward the cohort mean. Full method →

Logan Kilpatrick no published score: only 4 usable exchanges on raw tape, and a fair score needs 8+ · coarse estimate ≈4.0/5 from 4 raw tape exchanges record → ← everyone

Every exchange below was scored with names hidden, four dimensions each from 1 to 5. An exchange's score is 0.30·directness + 0.30·coherence + 0.25·precision + 0.15·compression. The published score averages the raw tape exchange scores and shrinks small samples toward the cohort mean, so five great answers can't beat twenty good ones. Produced feed rows count only toward coarse estimates, never toward a full score.

clear all ✕
4exchanges match
4on raw tape
0redirected or not addressed
Answered raw tape D 5 · C 5 · P 5 · Cm 4 4.85

Q And how expensive, you know, are we talking like, you know, in terms of, you know, tokens and, and what, you know, how, how, how does it, how much does it cost?

A Yeah, that's a great question. Um, so the fun part is that, uh, the model is available, especially through this experience of, like, you want to build an AI app or build an app in general. It's completely free. Um, so you can go to ai.studio slash build, access this, like, start building your product or experience. It doesn't cost anything, um, which I do think is, like, part of this headline story of making this, this capability accessible. There are limits, um, and after you hit the limits, we, Uh, we sort of prompt you to switch over to a, to a paid API key and then you can sort of keep going as far as you want and use it as much as you want. Um, the actual model is two dollars per million input tokens and 12 dollars per million output tokens. Um, so it's priced below some of those sort of comparable competitive models like, uh, um, GBT, 5.1 pro as an example, and Claude 4.5. Uh, is, is like lower priced than both of those models, and that's for, uh, input tokens less than 200,000, and then over 200,000, the, the price, I think it's like doubles for both, so it's four dollars per million in, um, and I think 24 per million out though, I'd have to fact check that. Um, so similar, similar pricing pattern to what we've done in the past for all these pro models, uh, which is great. It's adding a bunch of features, which is exciting. I feel like I'm going to go build this, uh, thi…

AI assessment note: “actual model is two dollars per million input tokens and 12 dollars per million”

Answered raw tape D 5 · C 4 · P 4 · Cm 4 4.30

Q much does this fire you up? Because, you know, uh, it fires me up when I see how powerful this is. To me, there's like probably a thousand business ideas that you can build on top of this. Could you, just cause you know the platform more intimately than I do. What's an example of applying this technology to a potential business idea that someone could go and start tomorrow?

A Yeah, that's a good question. I, I think that, like, um, the pulling out, like that, like furniture shopping example I like a lot, whereas, like, you have images, you have images that have, like, a lot of, like, really busy content, and, like, how do you break them up into, like, the right things, um, There's like, I mean, a bunch of like boring stuff, but I've worked with a bunch of companies that are maybe not boring for, for everyone, but like worked with a bunch of companies that do like inventory management and like this type of bounding box example, like matters a lot for them. So that like, they can just literally like snap pictures or have real time video feeds to like, you know, how much of the thing is being utilized. Like another boring example is like, uh, parking garages. Like you can have this like real time look at like the utilization of parking garages. You can sort of Do a very similar thing. Um, there's probably a lot of, like, meta stuff with, like, satellite analysis, being able to, like, really explicitly bound different areas, uh, based on certain criteria of, like, find me, you know, the cornfields and, like, make those, you know, and, and if you're trying to do some, like, meta analyses, um, those are all very random examples. I'm not sure how useful those are for people.

AI assessment note: “worked with a bunch of companies that do like inventory management”

Answered raw tape D 5 · C 4 · P 4 · Cm 3 4.15

Q What if you, like, okay, so let's just say you wanted to edit this image. Like, how do you actually edit it from here with natural language?

A Yeah, that's a great example. So in this experience that we're in right now, I think it's kind of hard, but let me actually just like take the image out of here, and then we'll go into another example. Um, so you could either, you have two choices. We could either just like go back to the main chat UI, um, and I could Dump the image in and start doing a bunch of stuff. Um, so we could try that first actually, and we'll see. Um, so maybe I say, make me a slow or let's do. Add the slogan. Pixel. The phone. For AI nerds. I don't know. I'm sure the pixel people won't appreciate it, but, uh, add the slogan pixel the phone for AI nerds, um, under the image, uh, of the pixel. And let's see what it comes up with. Um, So this is the, like, most lightweight version of this where you could come and do this. I, I do think there's something for, like, you want to have a little bit more, like, product scaffolding around this to kind of help you if you wanted to, like, you know, have a paintbrush to, like, highlight something or circle an area. Like, none of that exists in this default experience, but you can build all those things bespoke in that build tab that we were in before. You just need to, like, prompt and say, like, now add a bunch of, like, creative tools, and we actually have an example of this. Um, As well. But if we look at this pixel, the phone for AI nerds, um, it looks great.…

AI assessment note: “go back to the main chat UI, um, and I could Dump the image in”

Answered raw tape D 5 · C 4 · P 4 · Cm 3 4.15

Q You know, is there any best practices in terms of prompting to get, you know, the most out of, out of the product?

A Yeah, it's a great, uh, it's a great question. I think there's, um, and I don't, I don't know if I, I'll just throw another example up on the board or on the, on the screen as we, as we talk through this. I think some of the limitations, um, Are around the complexity of what happens when you do like multi-turn, um, When you ask for lots of edits in a single turn, I think the like best way to do this is to be like, like very precise single turn edits. The model is capable of like doing multiple things in a single turn if you're, if you give precise enough instructions, but I think it oftentimes like kind of loses the gist of what you were trying to say if you like layer multiple instructions together. Um, it is, it is also like, it's worth noting this model is powered by the, the same Gemini 2.5 flash model that like We released earlier this year. So it does have a lot of world knowledge. So you, you, you should assume it is like a smart creative partner in many ways, but just like any other, if anyone's ever done like, uh, uh, worked with like a random, you know, sort of creative, uh, contractor on the internet, like you oftentimes don't get, I just went through this recently where I was like, wanted someone to design a bunch of t-shirts for me. And I had sent them a bunch of stuff and I got the, I got the results back and I was like, yeah, this isn't what I wanted. And then I …

AI assessment note: “the like best way to do this is to be like, like very precise”

page 1
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 300 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.