AI Is Moving Faster Than Business Can Absorb It

Check out the full episode below! Enjoy The Exchange? Don't forget to tune in live Friday at 12 pm EST on the Greenbook LinkedIn and Youtube Channel!

What happens when the technology keeps accelerating, but the businesses expected to use it are still catching up?

That question kept surfacing in Episode 148 of The Exchange. Business leaders at Dreamforce are openly talking about how difficult it is to keep pace with AI. At the same time, investors are putting serious money behind AI search, research platforms are adding new ways to talk with respondents, and synthetic data is getting a harder look.

The technology isn’t slowing down. The bigger question is whether the rest of the business world can absorb it fast enough to make all of that progress useful.

Thanks to our producer, Karley Dartouzos. 

Use code EXCHANGE to get a 15% discount on your general admission IIEX tickets!

IIEX West 

Stay Ahead of the Curve! Subscribe to The Exchange Newsletter on LinkedIn Today! 

Transcript

Lenny Murphy: Are live. All right. Happy Friday, everybody. And Karen and I were just talking.

Karen Lynch: You have to hit. There we go. We're live.

Lenny Murphy: There we go. We're live.

Karen Lynch: Boy. Hey, Emma, you did it.

Lenny Murphy: Yay. Yeah. Yeah. Couple, couple disclosures for everybody. It's the one, my internet is iffy. If Karen, tell me if it frees up, I'll at least kill the video, which is probably a blessing for most people to not have to look at me.

Karen Lynch: Oh my heavens to Betsy, don't you go anywhere? Don't you go anywhere? Yeah, we're gonna hope that Lenny's video stays the whole time and his internet isn't um isn't crazy. So, you know, fingers crossed, it would be a gift for you to stay. Also, Karley attended a wedding in Europe, which you know is a lovely thing to say.

Lenny Murphy: So, such a pain, yeah, no, it's a pain in the ass, yeah, yeah.

Karen Lynch: So she is out today, and actually, she'll be out next Friday. So, Emma, our podcast producer, is producing this show this week and next. So, our process is going to be a little different. We're going to drop the links at the end of the show, just to make sure we don't mess up the live stream because we're not 100% sure of Karley's process. So, that means you can still comment, and the links are just going to be dropped at the end of the show. Emma and I were saying it. You know, at the end, we'll see how that flow goes for you. If you, if you like this format better, that the links are dropped at the end of the show, let us know. Um, because this is a good time to prototype a new process. So, so anyway, so those are the two caveats. So, um, there we go. So, there you go. There you go. Shall we begin? Let's begin. Let's begin.

Lenny Murphy: Let's have one plug? Last day for grit. If you've not provided for grit. Quite certain that you have probably gotten an email invitation or there, look on LinkedIn. There are tons of open links there. But help us help you participate in Grit. About 2,000 completes, which is fine, but the more, the merrier. So please, please, please join us. Any other plugs?

Karen Lynch: Yeah, we should plug AI next week. I mean, the registration numbers are looking great. Every time I check, there's, you know, there's like a couple hundred more. So we're looking fantastic for next week. We've got a full agenda. It's amazing. I can't say enough about it. You know, check it out. The sponsors are fantastic. The brands that are joining us are fantastic. The topics are great. They're just taking us to a whole new level. Get online, look at the agenda, it's free to attend. Um, I'm sure there's something for you. Register for the whole thing, join for whatever sessions you can. Um, yeah, I think it's amazing. I think it's amazing.

Lenny Murphy: So, we'll do a quick uh debrief next Friday.

Karen Lynch: Yeah, next Friday, I'm sure there will be so much to talk about. So, um, for real, yeah, it'll be cool.

Lenny Murphy: Yeah, so tune in, all right, and then uh, then West.

Karen Lynch: Right over there. Yeah. And then IAX West, we are, that agenda is full also. So that's very exciting. We're looking forward to that. We're in full-on execution mode, locking down all the production details for that one. That's amazing. So yeah, we're really psyched. It's going to be great. There's code exchange for 15% off general admin tickets. You know, it's just going to be a great show, and we're really looking forward to it. It's exciting energy to be in that region. I'm really psyched.

Lenny Murphy: So, late October in the Bay. I mean, I can think of a hell of a lot of worse places to be.

Karen Lynch: I know, I know. Yeah, it's really cool. I'm mostly excited for the startup energy, if I'm being honest. Like, I'm just very excited for just the whole startup vibe. And bringing the competition there. Yes. Yeah. And I think, I think, I don't remember if I mentioned it last week. You know, I've got all of the past five Insight Innovation competition winners from North America attending to sit on a panel and kind of talk about, you know, the founders, the founders' journey post-win. They're all in different places, and I just can't wait to talk to them about it. And, you know, it's right before the competition kicks off there. So. To make the finalists nervous, but to give them something to inspire them. So I'm excited.

Lenny Murphy: And have to say, this, you know, the last year or so, we saw a lot of clustering of types of companies in the competition. This wave, the finalists are differentiated, right? So we're kind of getting through the low-hanging fruit of the AI era of workflow automation, synthetic, online quality and now we're seeing some really interesting new applications. So I'm very excited for it. I love everybody who's ever gone through it, but it's cool that we're seeing some just more interesting things. I can't say it right. You know, just different stuff than we've been seeing. So very cool.

Karen Lynch: Yeah, very cool. All right. Excellent.

Lenny Murphy: Well, speaking in San Francisco.

Karen Lynch: Right? Yeah, like if you're paying any attention to the news in San Francisco, I'm sure it's local news. I've been reading some articles and seeing some video clips of local news. Dream Forces everywhere, which is, you know, sales forces take over San Francisco.

Lenny Murphy: Right. How many people go? I mean, it's an insane number.

Karen Lynch: It's an insane number. Yeah, Tim, my husband went one year, and you know, not only is every hotel booked, but the Overflow had him booked on a cruise ship. Like his hotel was a cruise ship, so it was a room on a cruise ship. So, right. That's how booked all the hotels are, which was quite an experience. So, yeah, really, really cool. But, you know, there's protests happening, not protesting Salesforce specifically, although I'm sure there's people doing that, but protesting AI in general. There's all the talk about the promise of AI along with the threat of AI. So, there's a lot of news coming out of that event. So, all you have to do is, you know. Google Dreamforce and click on news, and you're reading everything. You've tuned into some keynotes too. You know, it's a big deal. I did see one thing this morning when I looked at it. There was a headline that said, Business leaders say AI is already moving too fast for them. Business leaders at Dreamforce, AI is moving too fast for them.

Lenny Murphy: That's crazy. Let's set aside the political debates and all this other stuff. Because I realized the other day, just thinking through, whatever, it very well could be that the pace of technological innovation driven through the AI companies has outstripped the ability to absorb it. So. And not saying that it's not going to, obviously we're going to, but it just occurred to me that an argument could be made that the big frontier lab is saying, let's slow down.

Karen Lynch: Yeah.

Lenny Murphy: That may, there may be real, real legitimate threats there. Absolutely. Not going to set, not going say that's not true. But there's probably some business issues there as well. Maybe we've plateaued, there's competition coming in, we need to let the market catch up, you know, get some clarity on things. Open AI said, we're not going to IPO. So it's a very complex stew of things. But that idea of adoption lagging technology at that scale is an interesting idea.

Karen Lynch: Yeah, well, and if you think about it, yeah, if you think about it, when a business grows and they have to scale up, scale up too fast and too quickly, it can get unruly and out of hand, right? They have to be very deliberate about how they scale because that's a tenuous process. So, how is this any different?

Lenny Murphy: It's just magnified at an exponential level, it is right with massive amounts with massive amounts of capital. So, Anthropic does seem to be uh the rumor they're in IPO, and it seems like maybe it's firming up of a two trillion dollars the and and there's a fundamental question too. Like, we just had SpaceX at, you know.

Karen Lynch: To $2 trillion.

Lenny Murphy: And how much money is there in the market for these IPOs, too? There's just a lot of very complex, there's money at stake, lots of money at stake globally, not just here and this, these pragmatic. We may be just in that place where the forces that drive traditional business evolution. It's just too damn fast. And we just got to slow down across the board. Yeah. You know, for a variety of reasons, just so everybody can kind of catch up. And I get that. I get that argument. So I'm not sure if you're about the other ones, but I get that one wholeheartedly.

Karen Lynch: No, I do too. And I think on some level, there's something validating about hearing people who are playing in that ballpark talking about. About how fast it's going so that the people who play in our ballpark, we can barely keep up. Cool, because if that's the major leagues, then us down here in the minor leagues, we can be like, all right, we're okay.

Lenny Murphy: It is. Well, yeah, great point. Why did we start this? Why did we start The Exchange? Because there was so much coming, and we thought, boy. There needs to be a function to separate the signal from the noise because it's just non-stop. And yes, imagine the people that are actually driving, you know, driving the change and how they're feeling.

Karen Lynch: Yeah.

Lenny Murphy: It's just what an interesting world. Well, anthropic.

Karen Lynch: Speaking of anthropology, we just found out, and you know, not to, not to. Kind of go backwards in the program here today, but we are officially going to be welcoming Anthropic to our events next week. So they are joining their partner Nick. So, I mean, come on. Yeah. Sorry, Katja said T-ball. Like, it's true. I said we were in the minor league. So Katja, I didn't do T-Ball. Very cat. Very fair. I totally like elevated us a Yeah, so our partner Knit is bringing somebody from Anthropic. So we are literally going to have somebody from the major leagues on our stage. And I just got really excited about that. That's going to be, you know, first thing in the morning on, or, you know, one of our early sessions on Wednesday morning. So we'll have Anthropic, Anthropic. On the IAXAI stage. So, if you weren't convinced, five minutes ago when Lenny and I were just telling you you ought to come, now you should because, you know, that company's coming. So, I mean, we're excited for all of our sponsors, right? Quilt AI is coming and we'll be talking about them later today, too. And it's Speaker lineup, but yeah, Anthropic's going to be there. I'm just psyched.

Lenny Murphy: So, yeah, yeah. The companies that are creating the future in real time.

Karen Lynch: Yeah.

Lenny Murphy: Like hyper time. Real time may not be the right way to even think about this anymore. Yeah, they could be like, you know, five, you know, five hours ahead or whatever. Anyway, I won't go with my time travel analogy anymore. Okay. Lots of interesting billions of dollars. Speaking of billions of dollars, right? Let's talk about billions of dollars.

Karen Lynch: Yeah.

Lenny Murphy: What did you think about it?

Karen Lynch: Profound? Profound. So Profound raised $180 million at a $1.8 billion valuation to build their second raise this year. Yeah. Yeah. So building measurement around how brands appear in AI generated answers. So they started as an analytics platform, right, for marketers to understand how buyers are discovering or how buyers are discovering brands. Through AI. And, you know, now there's agents involved and they're doing some research and they're writing and they're reporting and they're delivering AI, AI marketer, and et cetera, et cetera. So they're doing a little bit more. So I mean, you know, this is the way. This is the way. This is what's happening. And it just, it, it, it builds on the story that we've been telling about brand discovery. And it's a new world out there, right? We're paying attention to it. We are tracking it and we are tracking how brands in our industry are being discovered. And we've talked about how important it is for people to be in our directory because our directory is visible. Et cetera, et cetera. But also brands out there in the global economy, how they need to really be considering how their brand is showing up. I mean, it's interesting. And this is a major capital bet on that measurement being really important.

Lenny Murphy: So they're obviously going to the next Nielsen or Concourse or Kana or. You know, they're going to become the standard of measurement in this particularly, incredibly important marketing arena. Yeah.

Lenny Murphy: Real quick, I was talking to, shout out to good friend Shane Skillen at HotSpecs. And Shane and I were talking last night because we're just buddies. And we were talking about this issue because they focus on brands. That's their whole thing. Anyway, And uh yeah, real thing brands are struggling to understand the role of the traditional brand, everything around your brand, uh, not just measurement, but you know, the attributes, et cetera, et cetera, in the world of AI disintermediation. And it is uh, this is a real challenge. So, the profound people are not, they're not betting on something that's that's you know. Imaginary. It is, yeah. First, how do we show up before we even get into how do we adapt, you know, our brand and our marketing and all that stuff? Uh, it's a big deal. So, yeah, yeah.

Karen Lynch: Hats off to them. Yeah, let's do the Brave Bison story next, because I think it's more closely related to this one. So, Brave Bison made another offer. I guess it made these offers before the offer for system one. You know, I think this is this, and this one is what, 47.5 million euros. What I think is interesting about this one is, you know, there's also, you know, this wheeling dealing in the shareholders or stakeholders, you know, how much money they already own. Like, they're already, they already have shares of the company. So it's a takeover bid. It's, yeah, it's, it's not that they're, you know, this is a, you know, a brand new situation here, but it's, it's the, the convergence of these two capability sets, right? So we're taking both the media execution and the measurement and just getting them closer and closer together because that's the play, right? That's the play. Yep, yep.

Lenny Murphy: Really, really interesting. Uh, and if I was a guy. System one was brain juicer, you know, just one of the storied companies in our space. Um, you know, really innovative and doing cool stuff. Um, yeah. Uh, and if you never had the opportunity to meet uh meet John Kieran, I mean, just what a wonderful man and the only Speaker ever to moon the audience. Um, he did not do an IAX, but he did a TED. Um, uh anyway. So yeah, yes, to your point, the convergence. Um, they're just moving closer and closer and closer. 100%. Um, the uh and then you get to the abstraction. Uh, you know, the companies that are struggling to be reinvented, um, the bending spoons, you've been following them. I mean, they've been buying, they've been buying up companies who uh at one point did really well, software companies, but have not managed the transitions particularly well. So they bought collaboration platform Miro for 1.36 billion.

Karen Lynch: I found that so interesting because 1.36 billion for Miro. Like at some point.

Lenny Murphy: Down from 17. But down from 17.

Karen Lynch: So, yeah, all right, all right. But it's a fire sale, yeah. Well, I guess, I guess, I guess that's the detail that really counts. But I still, I don't know, I still think it's a lot of money for what's basically you know, you know, a whiteboard, you know, a digital whiteboard. I mean, I don't know. As somebody who has used many, many times, you know, no offense, Miro, I think you're a cool platform. I used it for a lot of facilitation workshops. So it certainly has its place, but a limited use. So I'm curious as to how they would want to grow it and what. And I think that's probably why it dropped. But what's the plan? What's the plan? I guess I want to know the plan. And this doesn't talk about the plan. It says it's an AI innovation workspace. Okay. Right. What's the plan? Because it is still basically a whiteboard tool. It is still a digital whiteboard workspace for innovation. Yes. What's the plan to grow it beyond that? And I have to see what the plan is for it, because I still think 1.36 billion is a lot of money for something that I see as a limited use case. So, I don't know, bending spoons must know something. That's all.

Lenny Murphy: Well, they bought Domo too, right? They bought quite a few. So, there's, yeah, I guess we'll see. But it is. On a broader level, think about capital flows. And for our industry, right? uh, I mean, there are some stories, even you know, recently, Medallia and Ecent, and some of these companies that have lost value for a variety of reasons and valuation, right?

Karen Lynch: That's what I mean, yeah.

Lenny Murphy: And needed to retrench and be repositioned for future growth.

Karen Lynch: Right.

Lenny Murphy: So we have to assume that there, to your point, there's a plan.

Karen Lynch: There's a plan. What is it? Yeah. I don't know. Yeah, I don't know.

Lenny Murphy: I don't know either.

Karen Lynch: But here's how, let's segue into how people are growing. Some people who do have plans for growth that are. Are showing us those things week after week after week, right? So, you gov being one of them, you go is clearly showing us that they have plans for growth and they're innovating left and right. So, um, you gov expanding voices uh into their custom research. So, you know, AI moderated, moderated qual, not necessarily something brand new, um, but they're bringing it together with uh verified panel and quant data. So, um you know, kind of it's, it's when I Read this and I'm reading how they're leveling it up to what they announced the last time. I just think it's, it's the all-in-one, right? It's, it's their panel with their, with their AM moderated qual, and it's kind of, soup to nuts.

Lenny Murphy: They've been clear about their vision, being a public company, right? They've talked about it. You know, I am close to you,Gov, so I have to be, you know, a little. Conscious of this. But they're building the flywheel as everybody is trying to. But it is interesting that they took that EVA capability out of just the brand, the brand index, and now it's available for everybody and expanding that. And the most relevant, this, I mean, they were, you could say, a survey shop, right? I mean, fundamentally, that's what they're known for: political polling, et cetera, et cetera, you know, and the and built a panel to support all that and now shifting to have you know more of the moderated qual at scale and build a new platform, which we've talked about. We knew that we would get there, right? It was going to start replacing the survey to an extent. And there's a good signal of that. Yeah. Yeah. All right. We've got a couple things here. You want the two in one, the two in one. Yeah. Yeah.

Karen Lynch: Do you want to take it? Well, sure, right? So, Sirkana added AI-powered analysis and, you know, the good old, the good old weekly mention of NIQ with something called geo audiences. So the thing about the Sirkana story and this AI-powered analysis builder, the thing I like about this is that there's this visualization gallery. So, you know. AI-powered analysis may not be new. Visualization gallery is really what we're talking about needing and wanting in those dashboards. So that's what we want to be seeing. Anybody can talk about a dashboard, but being able to visualize that data is what has to happen on that dashboard. So I like that they just, they just got to right, they got right to the outcome of what that dashboard is for us. You know, they cut to the chase with their description. So I like that for them. The NIQ solution that they announced this time, you know, I was ready to be bored with their announcement. And I'm like, son of a gun, I'm not bored with their announcement. So I just, they did it again. This one is. This particular solution, this geo audiences, it's helping advertisers identify geographic areas showing stronger digital shopping activity. So, you know, like it's cool. They've just, they, it, you know, it's, it's, even if it's not necessarily earth-shatteringly new, it's a way to package what they offer by geography. And you know, bundled it nicely. Like it may be the same information that you can get from anywhere else, but they've now put a nice little package on it and said, Here, this particular segment that you've been asking for, we're calling it something. Yeah, and it suddenly sounds novel.

Lenny Murphy: So it does. And Europe. And Europe is a very complex match. Yes, there's the EU, but every country has its own.

Karen Lynch: Systems and laws, etc., etc.

Lenny Murphy: So, that's the benefit of a company like NIQ to get global scale, yeah, and knit things together. Uh, patchworks, what I was going to say earlier, you know, into something that is holistic and can drive, you know, because there's massively massive differences between Romania and France, yeah, uh, so yeah. Yeah, NIQ, keep on doing what you're doing. Sakana 2, they're showing up in the news an awful lot. And that is cool for companies of their scale and size. They continue to drive forward and innovate.

Karen Lynch: They're not resting on their laurels.

Lenny Murphy: So, very cool.

Karen Lynch: Hats off to both of them. Good stuff.

Lenny Murphy: Good stuff.

Karen Lynch: And another one. Yeah, take it away.

Lenny Murphy: Yeah. Informatic Target. They launched Buyer Intelligence. That is observed, and that was really interesting: observed person-level intent data. Particularly on B2B. So, there, you know, the informer tech target is going from data driving into activation. So, and to give purchase intent. So, again, it's synthesizing data at scale across. Lots of different inputs and making that useful to get to the bottom line. How do you grow your business? Who's your next customer? So, and they're, they're a very large company as well. So there we go.

Karen Lynch: Yeah. Yeah. So we got a couple of companies here that, you know, are just doing some good stuff with AI, right? So all in our ecosystem. You know, I'll take two, you take two. I'm talking about Outset. I'm really excited for Aaron Cannon. He added GPT Live voice-to-voice interviewing to the Outset platform. Aaron Cannon, who is the CEO and co-founder at Outset, posted on LinkedIn about their voice-to-voice interviewing powered by GPT Live. If you go to this post, which again, we'll show the link later, he's actually got a little demo like right there on LinkedIn. It's really cool because. It's just right in his post and it's very short. And I'm like, you know, I'm right here on LinkedIn. It's when I'm on LinkedIn, I'm in that kind of open-minded space. I'm like, I'll watch this short form demo. Great little clip of seeing how voice-to-voice interviewing works. I'm like, cool, well done, Aaron. I like what I see. And there you go. So hats off to you and what your team is doing at Outset. And pretty cool. Pretty cool use of GPT Live. So I was really psyched for that. And, you know, and then these other three companies, you know, I'll talk about Electric Twin because I'm really psyched because they were the, you know, they were the Expert Choice Award, which is, you know, kind of the grand prize, if you will, $10,000 check winner at IEX Europe. So, you know, I got to meet Ben Warner from Electric Twin. And I was just really excited about the fact that, you know, he's doing some great stuff with his launch there. So I know you can build on it with what the other two are doing also, because they're all kind of merged into this one big story.

Lenny Murphy: Yeah. Well, I mean, I mean, Electron was, you know, they predicted just its segue, right? They were working with TV and, you know, understanding advertising, et cetera. So with synthetic, with digital twins, effectively. Very cool. Now, original voices, who? Powers the YouGov synthetic, right? So, and they just announced a raise, it's a pre-seed of a million dollars to help scale their synthetic approach. And I had a chance to actually talk to Dave, the CEO, about two weeks ago. And just, you know, these are like two guys in a truck, basically. And not, I mean, in a disparaging way, it's like it's just amazing. You know, they're doing many cool things just leveraging these technologies in a new way to produce real impact. So they announced their raise. And then Quilt, the title sponsor of IXAI, launched their Anthropologic. And I mean, Quilt has always done just really. Interesting stuff to focus on kind of culture, et cetera, et cetera. And they do that off of open source data and run through the models and clean it and make it good. And now they've done that with synthetic. And I was playing with it last night. It's just a really neat way to be able to explore topics, particularly with a cultural lens. You still do very pragmatic stuff, ad testing, you know, et cetera, et cetera. It's optimized for that. But hats off to them.

Karen Lynch: Hats off to them. And I will share something very timely that, and so I, I have this, I have a LinkedIn newsletter that's my personal kind of personal, like a bit of personal writing. And I just kind of write when something catches my eye. And what caught my eye is. Something that an employee of Quilt wrote about last night. I was reading it, it was like 9:30 at night, and Michelle is her name. And she had posted something that kind of was culturally relevant, and it caught her eye. She kind of ran it through the Quilt program, and she wrote about it. I was like, Oh my gosh, brilliant use case for this platform. She put it on LinkedIn. I then wrote about it. So if you're curious, go to my LinkedIn page and you'll see exactly what I'm talking about. It is, an ad that is, you know, that is, um, that took up a lot of air time. And I don't want to get into it, but it took up a lot of air time this week for a lot of people. And so I was like, what a fantastic use case for a platform like that, where there is a cultural conversation happening. And they dug right in and it was, it was really cool to see it unfold.

Lenny Murphy: So, um, it is. Well, you know, it's probably one thing that's really that is all of these companies the recency of data is a component about the utility of the platform. And obviously the anthropologic quilt, they're pulling in real time from the web. The original voices is from the daily tracking data from yougov and et cetera et cetera and there is something to that um rather than a static you know trained off something from a few months ago it doesn't mean that's that's bad don't mean that at all but things do change to your point a new ad was launched people reacted you know uh it changes the conversation and that can be incredibly relevant to a brand who may have been ready to execute another brand or another ad just like that and thought oh wait Yeah, maybe not.

Karen Lynch: Yeah, yeah. Or maybe, or maybe that's exactly what they wanted. So who knows? But the reality is, it's, it's, it's pretty cool to see the ads. It just was really cool to see in this particular use case. You know, pretty cool to see that you can just. If you dig into these topics that are public, it's public information that they're analyzing. So it's really, you know, for curious minds, that was really cool to see. So I like it.

Lenny Murphy: Well, check it out. They were last on that, the Emperor Logic was on Product Hunt.

Karen Lynch: Yeah. So if you, and I think they had a deal for a free trial on Product Hunt.

Lenny Murphy: So go to Product Hunt, give them an upvote. And get the free trial.

Karen Lynch: Yeah.

Lenny Murphy: Yeah. All right. Let's do big tech.

Karen Lynch: We haven't talked enough about big tech. So, like, and just so you know, again, folks, we talk about the big tech because there's this trickle down to us, right? We only go there. Yeah. Yeah. So, um, so yeah, Google, Google released Gemini 3.8 live. And live extended thinking because we needed two versions. So one's built to be fast and cheap for high volume conversations. The other, this extended thinking, is built for harder requests. It's real-time speech-to-speech AI. It's voice interviewing, right? It's this voice-to-voice technology that we keep talking about, but it's the action that follows. It's the fact that, you know, I don't know, there's this, this, it, there's implications for AI moderation, there's implications for customer service, or, the customer experience, there's, there's implications for, for, you know, everything in our ecosystem, understanding infrastructure for what this can be doing. Check it out, check it out, check out what's possible now because uh, our future is going to be forever changed. Period.

Lenny Murphy: So, I have a confession. Um, I know the uh, I have started using voice prompts.

Karen Lynch: Um, uh, and I was like, I'm not gonna uh no, that's you know, you don't have to confess that you know, early.

Lenny Murphy: I do because I was like, no, I'm not going to do that. I'm not going to talk to them, you know, talk to my computer. But the efficiency gain from an efficiency standpoint to be able to, you know, just, I can be running four or five tasks at once and to be able to say, here's the, here's what I want. Here's the next step. Here's the and rattle that off from any device.

Karen Lynch: Yeah. Right. I know. I was just talking to Dan Wasserman at KGT this morning about how many tabs we have open at one time running different things and, and what all we're doing and how many, how many, anyway, how many GPTs we have doing things for us at once and agents are involved. And yes, so again, come next week. Dan will be talking by the way dan whilst I'm in a kgt will be talking about kind of uh you know the agentic future that we're stepping into on day two he'll be he'll be he'll be talking about what what's real for us you know he not the not not the major leagues he'll be talking about really what's what's real for us um and what what's what's coming in our future so um you know I think that's going to be another um another must attend session next week so anyway um yeah Yeah, back to Salesforce.

Lenny Murphy: And actually, both these are connected in my mind. Salesforce launched COA, their CRM reasoning model trained in their proprietary enterprise data if they code it with NVIDIA. What I took away from that was, you know, Salesforce, they're going for the infrastructure. You know, and off their proprietary systems. And yes, you plug in, they did, CloudForce a few weeks ago, and et cetera, et cetera, you know, they're doing that, but they're owning infrastructure empowering others, how their users can unlock value of their data. That is an ongoing piece of the debate we started the conversation with, right? Of the big labs and yada, yada, yada. So, you know, they're, they're, it to win it. But that reasoning model was interesting. And it goes to this next piece. I'm sorry, Karen, did you have any takeaway on that?

Karen Lynch: There's a quote in the piece that says: The most valuable thing Salesforce has built isn't our platform, it's the accumulated knowledge of how enterprise business actually works. And with COA, the knowledge is put inside the model itself. New on that is that that's what Salesforce has tracked the behavioral usage of the businesses. It knows how people have actually used it. And that's being put to work in COA. And I think that's what's really interesting to think about. So, you know, it's going to act according to things that will be comfortable. To things that are you know real workflows not just hypothetical or here's idealistic what you can what you can be doing but it's going to be very realistic um and going to be imaginable um because it knows it knows what people are actually doing in there and I think that's going to be pretty powerful that really is and that and so that that knowledge that context drives the reasoning the understanding and then this next one that TypeSafe came out of stuff with Jeb and is one of the, I co-creators of ChatGPT.

Lenny Murphy: But this is just about programmatic decisions, not a chatbot. So it's for decisions. So if we think about that flow of understanding, reasoning to a decision.

Karen Lynch: That's it.

Lenny Murphy: The evolution of these systems from an agentic is to streamline the workflow to get to a decision. And now we're seeing the emergence of these guys that it's chatbot style. If you haven't checked it out, look, there's all types of demos and people jumped all over this on X and they're all showing their work. And, you know, I actually at least got it. The GitHub repo to put it into Grockbot as well. But the point of it's just about decision making. It's just about processing a decision. Just boom, boom, boom. Without all of the re-rendering that the LLMs do, the traditional LLMs, there's no text. It's just straight to business. Reduced tokens. So just that efficiency now. We're seeing that focus on. The efficiency of what really matters: understanding reasoning decision, um yeah, and the cost is be massive cost savings and yeah yeah instant results, massive cost savings, yeah yeah, uh yes all right, uh Mr. Lynch, uh you all right, he's a fan of, uh type safe, um reinforcement learning for calibrated decisions, yes, uh R LCD, so back to the beginning of the conversation things are moving so fast. And not just the point of just the existing stuff, but where people are like getting inspiration and doing new things and the new, it's all shifting and accelerating in different ways, but all to create, unlock more value and impact across the board. So.

Karen Lynch: Yeah.

Lenny Murphy: All right.

Karen Lynch: Let's leave people with some further reading. So, this e-marketer story, we're going to let you kind of dig into it kind of links to the profound story, right? And the way search is changing. So, in e-marketer, it's reported that 41% of US B2B professionals use AI at work to start their research. And then verify what they find through traditional search. So they might get that, you know, that AI.

Lenny Murphy: Trust but verify.

Karen Lynch: So, anyway, so check it out. Think about what it means for your business in particular. And again, shameless plug. Think about what you need to be doing for your business to make sure you're getting seen.

Lenny Murphy: You know, let's give the shameless plug. You guys, I mean, I understand that, that, you know, the AI, they focus on quality of source and depth of information, et cetera, et cetera. And we see it. We see it absolutely. We show an individual company doesn't have the depth of information to necessarily compete against a directory, in this case, Greenbook. We see that the AI results, which are 41%, are going to AI first, but they're pulled from Greenbook. They're not pulled from your website. And that's going to be true for anybody else, right? So that, so recognize that is a shameless plug. And that's okay, because Greenbook is disagreeing about property. But more to your point, if that's the first stop is AI, where does AI pull from Where's the authority? Where's that? Uh, that is you have to think through that. The world of SEO, the traditional SEO, isn't necessarily the world that we're in anymore.

Karen Lynch: Um, and if a B2B professional is looking for something and they ask AI, and AI gives them an answer, and they're still going to check, they're going to check a source, you know, you want to be, you want to be where they're going to want to check.

Lenny Murphy: So, right. If the source came up in its Green Book, they're on Greenbook, they're not out for over 60 years.

Karen Lynch: I don't know. Maybe check the directory that's been around since it was in paper flopped off people's desks.

Lenny Murphy: Yep. Yep. I got a couple laying around somewhere.

Karen Lynch: Yeah. And then check out this last study. Final thing we're going to leave people with. Right? This is the last story, Emma. Anyway, good time to start dropping in some links. Flash Insight, you found this one. Take it away. I certainly can. Build on it, but talk about what you found.

Lenny Murphy: This is a need for study, I think. Laurent, he was the former president at Esmond, and he's an academic. So he's been building synthetic persona, publishing his results. So it was just interesting that they benchmarked their personal engine against 1,016 respondents. And just, it's the classic issue we see around dispersion: that if you're using off-the-shelf LLMs, they tend to cluster towards the middle. You don't see the nuance. And he's just documenting that, that there are a lot of differences. You miss a lot if you don't have a purpose-built synthetic solution that is compensating. So for first the right data, but then the right model. And that's. That's important. So hats off to him in showing up because it's not a substitute for human samples. If the real insight comes from the outliers, we're not seeing the outliers.

Karen Lynch: Yeah. Yeah. What I thought was cool was, you know, not only that they discovered a flaw, but they fixed the flaw, right? That it's like, oh, here's a flaw. We're going to fix the flaw. And then, you know, and then we're going to call out that without this fix, like, okay, there's a fix. Here's a flaw. There's a fix to the flaw. Without the fix, here's what's going to happen. Right. So it's kind of a call to action for everybody, like, you know, and see where you are in this equation, right? I don't know. I just thought it was, I think it's a pretty good call to action, like, you know. Because synthetic data is not going anywhere, right? It's just going to, it's going to get more and more and more dominant as we get out there for specific use cases. So if you're using it, you better figure out, you know, what's your correction for any of the flaws that are in the data.

Lenny Murphy: So that's right. That's right. Yeah. Yeah. I'll go to my soapbox just for a quick minute because I've been just digging into this. You know, it's not just do we have data and do we have an LLM to create synthetics. That's like the lowest hanging fruit, and it's also the least reliable. So it just is. So the focus is on the right data for the right answer to drive the right decision? And then is it the right model to your point that is calibrated appropriately to get as close as possible to how the human persona or the individual is going to base on information, and throwing more data at it doesn't fix it. You know, it's not a question of, you know, oh, you know, we have a million data points. Yeah. Okay. And are they the right data points? And is it being, is it, is the model interpreting it correctly? Yeah. So keep that in mind as you go through that, but it's an interesting world.

Karen Lynch: Yeah. It really is an interesting, interesting world.

Lenny Murphy: So it is all right. You have a baseball game to get to.

Karen Lynch: Oh, well, it's not, it's not till tonight. So I was saying to Lenny folks, tonight, um, tonight I will be going to a Mets game, and you can tell the real Mets fans will be showing up when you know that the Mets are not going to make the playoffs, they've had a horrible season, um, they've had a horrible week, and if you know, you know. Know. But some fans still show up, right? That's how you can tell a fan from a, you know, I don't know, from just a person who likes a baseball game. Like the real fans still show up, even when there's no chance of the playoffs and it's been a terrible season. We're still going to go. We're going to have a good night.

Lenny Murphy: Well, I hope you and Mr. Lynch have a good time.

Karen Lynch: I hope we will.

Lenny Murphy: Drink a beer, eat a hot dog, you know?

Karen Lynch: Both of those things are possible. Although at City Field, there might be a lobster roll. It is City Field. I don't know if you know that about City Field.

Lenny Murphy: I didn't. I don't know. It's a lobster roll in a stadium in the middle of summer. I don't know. That does not sound like a good recipe for me.

Karen Lynch: You would be surprised.  Okay. Yeah, no, it's actually, and it's, you know, it's our kind of lobster rolls, the hot buttered ones. So, um, and by our kind of, I mean, the Connecticut lobster rolls. So, they've done a good job with that, not the, not the, not those cold lobster rolls with mayonnaise, you know.

Lenny Murphy: But I despise mayonnaise. I'm glad you said that because I actually didn't know that there was a difference. So, if you say a hot buttered lobster roll, that catches my interest because I was. I'd never eat a lobster roll because I only knew about the cold mayonnaise once.

Karen Lynch: Oh, yeah, yeah, no, up here, all the hot buttered lobster rolls. So, and you know, apparently in the New York area, we know, we know the truth up here. The hot buttered lobster rolls are the only way to go. But there are, when you travel outside of the area, sometimes you don't get so lucky. And they're like, What are you even doing to the lobsters?

Lenny Murphy: So, there's one last thing. Yeah. Happy birthday to you. Happy birthday to you. Very, very Happy birthday, dear Karen. Happy birthday to you.

Karen Lynch: Kat, thank you, by the way. I really appreciate that.

Lenny Murphy: You are. You're welcome.

Karen Lynch: Thank you, Katie. I was thinking, Kat.

Lenny Murphy: I knew I was missing. No, thank you, Kat. Yes, I'm thinking of Kat, too. You're not thinking of me. It's like, oh my God, you just ruined my birthday, Lenny.

Karen Lynch: No, thank you, Lenny. It's a gift that I can hear you sing.

Lenny Murphy: I will take that exactly in the spirit that it was given.

Karen Lynch: Happy birthday. hank you. Thank you very much. Thank you very much. Yeah, no. Birthdays are fun, right?

Lenny Murphy: They're fun. Yes, they are fun. And thank you, Kat, for reminding me that I knew I was missing something.

Karen Lynch: I knew it. So that's great.

Lenny Murphy: All right.

Karen Lynch: So go and join. Next week. Thank you. And yeah, be prepared to talk about IXAI next week because it's going to be hard to get to the news if I've a lot to talk about.

Lenny Murphy: Yeah, absolutely. And let's not, Emma, when we post this, let's not say. And wait for Lenny to sing at the end. We want people to actually come.

Karen Lynch: Yes, we will not have an encore.

Lenny Murphy: No, no encores, and do not advertise that. I am keenly aware that I have no musical talent and can't carry a tune in a bucket. So, anyway, happy birthday.

Karen Lynch: Everybody, take care. Have a good weekend. Thank you. Thank you. Bye, everybody.

Lenny Murphy: Good to see you all. Yep. Bye.

Karen Lynch: Bye.

Links from the episode:

Bending Spoons to buy Miro in $1.36 billion cash deal 

Brave Bison made a fourth offer for System1 

YouGov expanded Voices to custom research 

Circana and NIQ are expanding their AI and audience tools 

NIQ Launches Digital Purchase GeoAudiences Across Europe 

Informa TechTarget launched Buyer Intelligence 

Outset added voice-to-voice AI interviewing 

Electric Twin tech used to test TV ad with simulated audience 

AI Data Firm OriginalVoices Raises £1M Seed Funding 

Google introduced new Gemini 3.8 Live capabilities 

Salesforce and Nvidia introduced Koa 

TypeSafe introduced Jev 

EMARKETER reports that 41% of U.S. B2B professionals using AI at work start their research with AI before verifying through traditional search. 

FlashInsight found that its synthetic consumers agreed too much compared with real respondents