Digital Twins, Synthetic Data, and the New Rules of Research

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!

AI is rewriting the rules of research and marketing, and not always gently. This episode digs into what happens when digital twins start speaking for us, when synthetic data gets too convincing to question, and when agents start making decisions humans used to make alone. Big acquisitions, bigger implications.

Thanks to our producer, Karley Dartouzos. 

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

IIEX West 

IIEX.AI 

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

Transcript

Lenny Murphy: Hello, everybody.

Karen Lynch: Hello, hello. I know. We just, full disclosure, it's a funny start because we are not live. Well, we're alive.

Lenny Murphy: We're alive.

Karen Lynch: So that's a spoiler alert. We are not synthetic yet. Not yet. Not yet. Anyway, there are no digital twins of Lenny or I yet. We are pre-recording today because there is a conflict tomorrow and tomorrow, which is today for you all. So we're recording Thursday, August 27th. This will air midday on Friday, August 28th. So we're only like, you know, 16 hours ahead of you or something. Yeah, not 16, that's bad math. A couple hours ahead of you all. But happy Friday to you all. And I. I hope you all had a great week. There's a whole lot that's happened since I saw you last. How did it go with Sarah? I did not catch up because I was on a beach.

Lenny Murphy: Well, of course, it was great. We had a lot of fun. Uh, we did set the record on the longest, uh, the longest exchange. So, just shy of an hour.

Karen Lynch: Oh, gosh, okay.

Lenny Murphy: Well, yeah, uh, so I don't know if that's a challenge or what, but the uh, um, but it was, uh, it was fun, good, all right. You look, you look wonderful.

Karen Lynch: You look, you're tanned.

Lenny Murphy: You look relaxed.

Karen Lynch: So, thank you. Thank you. You know what? It was a 10-day vacation on the beach, which I mean, I didn't sit on a beach for 10 days, but it was 10 days in a beach community. And I read three books, which was amazing. So that's how unplugged I was. Three fiction books, too, not even business books. I just unplugged. I saw all of my children and my grandchild. Everybody came to visit us and stayed for some portion of the time. We brought our dogs with us. You know, it was an Airbnb that was sort of come one, come all, come anytime you want. So it was to say it was the exact kind of break that I needed. The exact kind of break that I needed. You know, you don't really know, and this is sort of a PSA for everybody. You don't really know if you're approaching burnout because you're overworked until you get away for a while and you're like, oh, I was dangerously close. You know, when you realize kind of the pendulum swinging the other direction and you are like, oh, I really needed that. I really needed the break and I feel radically different, you know, in a good way, in a way right now. So I feel great. Good.

Lenny Murphy: I'm glad. I'm glad. Well, shall we, let's do our shameless plugs?

Karen Lynch: Yeah, because there's a lot, there's always a lot going on. But I feel like we do talk about these events all the time, but they're so important. We have to share them with the world. They are.

Lenny Murphy: Well, and let's start with sharing with the world, especially. Grit, so help us help you. The lace grit survey is out there and please participate. We need your input because there is a lot happening, as we've talked about many times. The pace change is now measured in weeks, certainly not months or not years. So, yes, you just took it a few months ago. Yes, we need you to take it again. So, we can. Understand where things are going, and we will deliver all that information back to you here in the next couple of months.

Karen Lynch: Yeah, but also I think it's important to point out these two reports, there are two each year and they do track different things. So even if you feel like I just did the grit report, you did just do the grit report. We have, there's sort of six months between the iterations and one is kind of based on insights practices and kind of the work that we do. And one of them is really kind of business innovation. So slightly different, similar, similar, but different?

Lenny Murphy: Yep, yep. The sequences we go to field Q1 and Q3. That's the sequence we publish in Q2 and in Q4. So, yeah, particularly interested in this one because there's some questions that we've been tracking. Adoption. Boy, wait till this, you know, all these topics, synthetic adoption, blah, blah, blah. Right? Yeah. This is where we measure. It's not just here, we talk about the buzz, right? We, what's happening, what's happening in market movements. That's where we get to what people are doing.

Karen Lynch: So, yeah, please. And speaking of what people are doing. Well, yeah, but real quick, I just want to say I have seen a lot of our partners putting it on LinkedIn. They're sharing it out there, encouraging people to participate. And I have seen it and I have noticed. So thank you to the partners doing that. I'm sure you've all gotten emails to say, you know, kind of put it out there. So thank you to those of you that are doing that. And if you've gotten the email that says, hey, help us spread the word, please do that as well. It's super helpful. So.

Lenny Murphy: 100%. 100%. Yeah. So, cleverly segue into IIEX.AI in terms of what people are doing, right? Real case studies.

Karen Lynch: Good segue. Good segue. Yes. Yeah. I mean, this event is going to come up very, very fast because now we're less than a month away. And, you know, our registrations are. Going to the peak. We're well over a thousand registrants now. It's free to attend. So there is, you know, there is no reason why you should not show up, learn from some of the best in the business, from stellar brands, from amazing partners, you know, really showing tangible use cases from how to build a research. Work stack for a researcher for your own purposes to how you are actually using synthetic research effectively in your research repertoire. So it's covering the full gamut. So it's really going to be a brilliant event. I'm excited for it.

Lenny Murphy: It will be. And then, so that's our virtual event, and then our new.

Karen Lynch: Entrant into in-person. Yeah. Yeah. This one's this one's, you know, I feel like a broken record every time I talk about this event and how excited I am for this event. I, I, every time I talk to a speaker about their topic, are checking in with me to make sure that, you know, they're on track. It's just amazing. It's going to be, yeah, it's going to be amazing. So check it out at the very least. Get yourself there. If you want to get there and you have questions because you've never been to, you know, if you've never been to IIEX and you're listening to this show, that will surprise me. But certainly reach out to us, the exchange at greenbook.org. We can tell you all about it. We can probably get you a code too.

Lenny Murphy: Um to to come as a first timer so just let us know yeah absolutely uh you may have just noticed we have a special guest um today this is prison he's a kitten who's decided that he wants to lay on my desk and uh and you know how cats are I keep putting him down he keeps jumping up so if you see prismo jumping into the frame uh he's doing what a kitten what a kitten does, so anyway, I love all right.

Karen Lynch: I just think it's so funny that you are quickly going to become the crazy cat man, dude.

Lenny Murphy: The crazy five kids, three dogs, three cats, four chickens. Come on. I've been crazy for a long time.

Karen Lynch: That's for damn sure. But I just think you, there's just something about you being surrounded with cats that's just bringing me some joy.

Lenny Murphy: So I actually, I, I. I am more of a cat person than I am a dog person. It's a fine line, but my whole life I've always had cats. So I love cats.

Karen Lynch: No, same. We have a cat also. We are cats and dogs as well in this household. Not as many as you. We are a one-cat household. Two dogs, one cat. So anyway. Yeah. But enough about that. We've got stuff to talk about.

Lenny Murphy: Let's do that.

Karen Lynch: Like, there's no segue to go from cats and dogs to NVIDIA. So, like, what do we do there?

Lenny Murphy: No, well, fighting like cats and dogs for, you know, anyway, the uh, it is interesting. NVIDIA. So, yeah, two, uh, two, uh, two deals with uh NVIDIA this week. Um, uh, they 've talked about the ecosystem changing, it's focused on infrastructure. There is no more fundamental infrastructure player than NVIDIA. And now they are reportedly buying Hugging Face at a 13 billion valuation. Hugging Face is like three years old.

Karen Lynch: Yeah.

Lenny Murphy: And then also invested in Mercor at 20 billion, which is human-generated AI training data. So the NVIDIA. Speaker. Well, nope, they have their own model, Nematron, which if you haven't tried the Nematron model, it's really good. I think it's almost as good as any of the other standard models. And now they're building their stack.

Karen Lynch: Yeah. So, and I think we said we talked about this a couple of weeks ago. Like, these things don't really get leaked unless there's some truth to them. So this, it's at this point, if they're if these names are being, you know, they're, they're, this is, this is, there's a strong probability, would you say, if these are being leaked at this point?

Lenny Murphy: I, I would. I mean, you were talking beforehand, right? There's rumors. We're, we're always sitting on a lot of rumors. Uh, the, the discipline is we don't say those rumors publicly until there's been a published leak. Yeah.

Karen Lynch: Yeah. Right. So, um, what's cool is the, the, one of the, I the link we're sharing is um, there's a LinkedIn conversation happening around the second story about investing in Mercury, um, Adrian from Mirsch. So, he is a 2026 future list honoree friend. So, check him out, check out this post generating quite a bit of conversation about this. And, um, so you know, go head on over there. I love, and I think Lenny, Lenny, you do as well, like, love when the community. Get engaged in these types of conversations because, you know, we at Green Book take our mission pretty seriously about helping to drive the industry forward and kind of ushering in the future of Insights. Our tagline, after all, right? And, you are strongly about it. That's part of how this company, you know, kind of began. And I think that. When we see people engaging online in that way, it gets us both. Like, we're like, ah, that's what we like to see. We like to see engagement in the community. So anyway, head on over to that post and check it out because, you know, this one in particular is, I think, you know, just an important one to check out.

Lenny Murphy: Yep. Yep. Another follow the money, another week, another synthetic.

Karen Lynch: Yeah, but this one gives me the creeps, though. This, this one is, I don't know. This one is because it's talking about employees, synthetic employees.

Lenny Murphy: Anyway, well, it's but building your digital twin was my take. So, like back to our previous, the beginning of like, are we real yet? The twin won 20 million in seed funding. Yeah, seed funding. Yeah, that in mind. That probably means they are pre-revenue or just a small amount of revenue. Yeah permission ai powered professional digital twins now this you know steve phillips a few months ago built the uh ceo friend right which is basically the persona of you know and we've talked about like what is can we build a digital lenny right yeah but this one but this one this one if you go into this this um and this story about it it can communicate on your behalf via email meetings, documents, Slack, Microsoft Teams.

Karen Lynch: It's trained so well in you that it can think of you as an AI replica of you. That's why it's giving me the creeps because that's how confident they are that they would know how to respond in a meeting.

Lenny Murphy: Yeah, yeah, I hear you.

Karen Lynch: Do I want to do it, do I not? I even see the Slack response, and then I'm, and then I'm, and then I'm in another meeting, and I don't know how I responded in the Slack response. I don't this technology in this context raises questions for me, and I understand that it's being funded and it's probably going to happen, and we're going to see it in the near future, but how do we as humans handle that tech development when I don't know what my replica has said on my behalf.

Lenny Murphy: I hear you. The, I, you know, I know you've done this as well. I, I, I know that I'm like two steps away at this point of having kind of a second digital brain, so to speak.

Karen Lynch: Right.

Lenny Murphy: You know, I've set things up with my use of AI and, you know, and it has a lot of context and blah, blah, blah. And, you know, it knows my voice. And, you know, for the most part, except for the end dashes, right? But the point is, it is pretty close, but I will not give it permission to do it, even though it asks all the time, right?

Karen Lynch: Would you like me to send this email? Yeah. Yeah.

Lenny Murphy: You can write a suggested email for me. I won't give it that permission. I won't. I'm not there to give it those things or to let it even create a fulsome, comprehensive version of me because it is weird, even though I could see the benefits. I could even see the benefits from a monetization standpoint. Right. But yeah, I'm not willing to go there yet.

Karen Lynch: So, yeah. Permissioned. It does say it's permissioned, right? So, there's questions around that, how it's, and then there's, it requires governance, right? How it's how it's, how, how is this going to be, um, how is this going to be, you know, rolled out in an organization, right? Even if I'm granting it permission, then how does it report back to me? And I can imagine there would be times, even if I give it permission, I can't attend this meeting, please act on my behalf. I can imagine there'd be times where I'll read everything that it said, and you know, but hopefully it did the right thing by me. But then other times I might just be too busy and I couldn't read that, you know, then I don't know what was said on my behalf and I missed the memo. So. Anyway, we'll pay attention to the thin one. Yep. We'll see what happens next. But yeah.

Lenny Murphy: It's the work world. It is. Now I was involved in this next. I was working with Brado, Brado Digital and Insights. Yep. As they were going to market. I was not involved in this deal with merge. So the cool thing about this was. That this is a little different than some of the stuff we may touch on here.

Karen Lynch: The um because Brado is a very human-centric, qualitatively grounded consultancy, and so taking a step back for the audience, they acquire, merge acquired Brado, acquired Brado Digital, and inside practices, right?

Lenny Murphy: And merge a more healthcare centric kind of advisory consultancy, uh data centric, but they wanted the human depth component that Brado brought. So I really love this because so much we talk about the tech, tech, tech. Brado is anti-tech. They, you know, sort of use tech in but deliveries of things, but they're very human-centric. So I love that we're seeing now the other end of the spectrum, that it's not just pure consolidation. You it's also the human expertise, human depth, you know, and acquisition, you happening around that as well. So hats off to the Broado team and to the Burge team. I think we'll see more of these specialty consultancies, human-driven consultancies. Now being acquired by larger companies because they bring not just business and relationships, of course, that's part of the equation, but they bring that missing element that the twin-one AI isn't going to be able to duplicate. So yeah, so that was kind of cool.

Karen Lynch: Yeah. Speaking of things we see more of

Lenny Murphy: Yep. Yep. You want to.

Karen Lynch: Guess who's in the news again, friends? And if your first two answers aren't NIQ, your third one needs to be.

Lenny Murphy: That's right. That's right. I mean, for real. Or listen.

Karen Lynch: I mean, exactly. Exactly. And the Open AI Deployment Company announced a collaboration bringing consumer intelligence into enterprise workflows. Basically, there's like two different products they've announced. They're extending and deepening their AI product capabilities. Consumer intelligence is really the place to be. They're all about where decisions are happening and making it accessible. Less about reports and more about kind of, you know, workflows. Yep. So. Here we are. Yep. Once again, their PR machine. Like, yes, they're, they're, they're shipping lots of products, but also their PR machine manager, you know, but they're on it.

Lenny Murphy: The OpenAI deployment company that was launched, what, two months ago, right? That was the consulting implementation, the forward deployed engineer, you know, type model with whatever gajillion dollars, you know, put behind it. Right on it, wait, our data can be integrated. You're looking at building these custom infrastructure deployments of AI organizationally. We want to be in the mix.

Karen Lynch: Yeah. Stellar team over there, man. I mean, I think, you know, if I were hiring, I'd be thinking about looking at like, who are the people on these teams? Because they are doing it.

Lenny Murphy: NIQ's quick ass. They really are. They're redefining. The entire business and the category. That's pretty cool. Now, the flip side of that, right, kind of the opposite direction of this next one, Salesforce and Anthropic, Claude Force, sorry, not force.

Karen Lynch: Claude Force. Yeah.

Lenny Murphy: All the Salesforce workflows and data directly into the cloud. So MCP. So. You know, again, same, same thing, same model of how you integrate data and bring it to customers where they are.

Karen Lynch: I think there's a line in here that says you'll never need its app again. Like, whoa.

Lenny Murphy: For a SaaS company, that's a pretty big fucking deal.

Karen Lynch: Bold statement.

Lenny Murphy: That's a bold statement for a SaaS company, right? To say it is not about the interface. It is about the functionality and the underlying.

Karen Lynch: Utility of the data.

Lenny Murphy: Yeah, that's a, yeah, I almost stopped the NF bomb.

Karen Lynch: Yeah, yeah. Well, and think about that. If you are a research platform, think about that. Think about that pivot for them and what that might mean. Think about how bold that is and how confident they are that Salesforce is saying that and doing that. That's a confident, bold move. Yep. What does that mean for your business? And are you? Can you make a bold move? Because that's right. That is fascinating. I was like, whoa, amazing. Amazing. Absolutely.

Lenny Murphy: Big, big deal to realize is the value of a company based upon the UX on the interface for the last 30 years. That was the model.

Karen Lynch: Now it's not necessarily. I know. I know. There is a woman from Salesforce speaking at our AI event, by the way. So, you know, come to that event. She'll be there. And maybe there's a great place for questions at the end of our sessions.

Lenny Murphy: Yeah. Yeah. Very cool. A whole bunch of new product launches.

Karen Lynch: Oh my gosh. Yep. Yep. Speaking of synthetic. Yes. Yep. Yeah. Yeah. You want to go there? Well, Zappy, I mean, Zappy launched synthetic idea pre-screening. It's a perfect use case for, we've talked about a great use case for synthetic pre-screening ideas. So early stage innovation. So, you know, it's not necessarily a surprise use case. It's the right use case, though. So, you know, you know, good on them.

Lenny Murphy: Absolutely. Yeah. So cool. Cool for them. And, you know, go back to IIEX, of the very first competition winners, not the very first.

Karen Lynch: It was one before them, but the second.

Lenny Murphy: And look at them now. Her baby's all grown up.

Karen Lynch: Yeah.

Lenny Murphy: This was the system. One launch. There's a little more to that. Did you dig into that one?

Karen Lynch: Well. Yes, you know, so system one launched to test your ad screen, which is an AI tool for screening up to 100 long-form creative assets at once. So the digging in, you know, not necessarily, Reddit didn't dig further than that.

Lenny Murphy: Well, so it's. Yeah, it's sorry, I didn't mean it sounds like a test, but yeah.

Karen Lynch: No, no, no.

Lenny Murphy: When I first started with, well, yeah, okay, they're like, okay, they have a normative benchmark or normative database and they're scoring Okay. They're closer to synthetic. They won't, they didn't use that terminology, but that was, this really is getting closer to, I would call it a hybrid between kind of a normative, normative benchmarking. Database and a synthetic, and yeah, uh, to be able to test that at scale and long form, long-form assets, not just your you know, traditional ad. So hats off to uh to yeah there’s two out of the UK zapping and system one doing some cool stuff. So, yep, yeah, pretty cool. You want to talk about M4?

Karen Lynch: M4. Yeah, so M4 launched something called Danny D-A-N-I, which is adding video interviews and AI analysis to their quant studies. So basically, at a certain point in a quant study, they can simply ask respondents to turn on their camera and explain their answer. Like, oh, tell us why. So kind of halting the quant study and say, camera's on, explain your answer. Um, so you know, a little bit of a little bit of qual in the middle of a quant survey, richer, richer answers. Um, just folding a little bit of real-time video input. Um, I think it actually solves a couple problems at once, right? It's gonna, it's gonna, it's gonna solve a data quality issue because bots will fall off at that point, you know.

Lenny Murphy: Well, which they would already solve. I mean, I, I, I will. So hats off to M4. I mean, they've always been a little counter-revolutionary on this. It is an app, it is a behavioral app. Their panel is very high quality. They're capturing behavioral data, they engage in research. Also, bear in mind, they got a significant, if not majority investment from SEMRush. Which is now owned by Adobe.

Karen Lynch: Yeah.

Lenny Murphy: So the data throughput and the way to augment the data now with video.

Karen Lynch: Yeah.

Lenny Murphy: That's a pretty, pretty cool flywheel.

Karen Lynch: Yeah. Yeah.

Lenny Murphy: That they've put in place there. So because video is the, yes, you're right. It's the data quality component. For them, it's maybe not supposed to be a validation layer because they already are. They kind of have that already by their very nature, but to take it further, so they're getting the richness, they're getting the richness, they're getting more the richness, absolutely. Um, so hats off to uh to them. Uh, uh um Phoebe, another winner of the Instagram competition, yeah, launched their and they did voice analytics before voice analytics was cool, right? This is in 2020, so. Um, uh, they were an AI voice analytics company when they won the competition in 2020. Uh, uh, now they basically turn everything into a telephone interview with Katie, right? Yeah, uh, regardless of mode, uh, doing voice. So I think that's really cool.

Karen Lynch: Um, yeah, yeah, and this one says it's voice, it's the voice technology combined with some behavioral science combined with AI to reduce that Katie time. So a little bit of everything mixed in there. You know, so their tech is really modernizing those phone research ops. So good on them. I'm sorry.

Lenny Murphy: Look at that. What happens when we're this?

Karen Lynch: Happened the last time we pre-recorded.

Lenny Murphy: Because I'm not thinking the right thing.

Karen Lynch: That's right. So funny. I was like, that only rings, you know, on Thursday afternoons.

Lenny Murphy: It does not ring on Fridays because I silence it. I have a whole preparation. Funny. I really like it. I don't have a cat on my desk or my phone on.

Karen Lynch: Anyway, so all good things. All good things. All right. Check.

Lenny Murphy: Yeah. So we. You know, you look at the corpus, we could have had like 20 things on MCP launches. Like literally. So that was the thing this week. Everybody launching MCPs. So I really like how you just picked. Well, let's talk about the model level, right? New standards emerging for model context protocol, the roadmap.

Karen Lynch: I think this was something that you had found. I didn't find or, you know, you Well, but you picked it. Yeah. You picked this.

Lenny Murphy: Well, the bajillion MCP.

Karen Lynch: Yeah, because it's a summarization, at a high level. You know, that proves oh my gosh.

Lenny Murphy: Oh, see, guys, this is what happens when we do, we figure we're both.

Karen Lynch: No, you don't know. Everybody, stop.

Lenny Murphy: Look, it's the end of summer, guys. Bear with us. The last Friday in August.

Karen Lynch: We're almost into September. Oh, my heavens.

Lenny Murphy: Oh, my heavens. So, everybody thinks that the MCP is not a thing. Well, it is a thing.

Karen Lynch: Yeah, it's a big thing.

Lenny Murphy: And it's maturing. Stop.

Karen Lynch: So, again, I'm turning it off. It's going off.

Lenny Murphy: All right. There is also a whole bunch so this next piece on cloud fair yeah cloud flare sorry bot preference sync yeah to block opaque ai crawlers from sites that disallow training there are another what five or six different articles related to this more than that actually when you get into the uh kind of the publishing and and uh agentic disruption component another huge theme this week yeah um But Cloudflare, right? Think about that. They're an infrastructure player. They're like sit, one of the critical, critical companies for the entire internet, really. Yeah.

Karen Lynch: Yeah.

Lenny Murphy: And this blocking content is becoming a significant thing. So the battle over consented data. Really is now moving absolutely into the website ecosystem. And we could have a whole other conversation about this.

Karen Lynch: Yeah. And I think, you know, we plunked these two stories here this week, you know, not because every insights professional is listening to them is saying, like, you know, this is really relevant for my job, but if you're you're the business owner and you have developers on your team, for example, these are conversations you need to be having with your dev team or your engineers on your team. These are, you know, just go talk to them about MCPs or go talk to them about preferencing. Like these are big stories right now. Go talk to them. I'll transform conversations.

Lenny Murphy: Yes, but the bot preferencing component, there are so many, think about companies. Social media analytics back in the day. Now, the kind of open source synthetic platforms that are training off of the web.

Karen Lynch: Yeah.

Lenny Murphy: Someone was describing a new business to me earlier today that sounded really cool. It was an iterative innovation process, but the, but they keep going back out to the web to get the data. And my first thought is. One look at this like Cloudflare that may be coming to an end, yeah. Um, at least for free, yeah, two so much crap out there that already is AI generated. I mean, the days of us are just changing, and there's an example of that.

Karen Lynch: So, yeah, and I mean, the value of opt-in or permission or consent-based human data is just going to go up. So if you are a business that has that, treasure it, treasure it, keep it safe because that is becoming an asset.

Lenny Murphy: 100%. Built. It is a moat. It is a moat and an asset. So, all right. These. A couple read more. Yeah, some good reads.

Karen Lynch: I mean, some good things, like this, weren't even all of the reading material this week. Like people were chatty and like, I just was like, all right, who are we going to call out for the sake of time? Because we are not going to give it an hour-long show, right? But these were the four I was like on theme and, you know. I would just say these are the four.

Lenny Murphy: Team up. I Team up.

Karen Lynch: All right. Finn, Finn, Finn Robin talking about synthetic data risk. Like, but, you know, again, these are on LinkedIn. Go on there, have this conversation. He's talking about how synthetic data is becoming more convincing. You know, I think one of the quotes that I wrote down here is, um, is it becoming less meaningful? I think he wrote the better question: is, can you demonstrate why I should trust it? Right. So that is a great question for those of you who are in that business: it demonstrates why I should trust.

Lenny Murphy: Yes.

Karen Lynch: Because it's plausible. Yes.

Lenny Murphy: Yes. This was the last week we talked about the whole medium strategies. You know, polling, op. It was all just BS. We don't need to rehash all that.

Karen Lynch: Go back and listen to last week's anybody. Yeah.

Lenny Murphy: But it, yes, validation and calibration. I think that it's always been important. Now, in this world of synthetic and AI, it's even more important. Finn does a great job. Arguing about that and tying it into this because it has a consequence. This whole media and strategies thing, I mean, they were trying to sway the outcome of an election with bullshit made-up data. So, but it passed the sniff test, the initial sniff test. Yeah. So, even though it had, it was utterly fake.

Karen Lynch: Yeah. Yeah.

Lenny Murphy: Um, so there's real consequences to uh to that. So, yep, I thought it was good.

Karen Lynch: Yeah.

Lenny Murphy: Our friend Jim Whaley.

Karen Lynch: Yeah. I mean, I saw this one came across my feed, and I'm like, oh, look at that. Jim Whaley's writing for Forbes, which, you know, I love. But his argument in this article. So, yeah, great article. Good job. Arguing leaders should treat synthetic data and traditional market research as complementary rather than mutually exclusive, exclusive, which. You know, landing in our laps, that might well, yeah, we've been talking about that, but this is in Forbes, which means a broader audience. So that's why I was particularly pleased with it. And there's some, there's some great quotes in there. Towards the end, he says here, leaders and their teams should adopt the mindset that it's a great day for discovering why, which I like, I love that. It's a great day for discovering why. Yeah. Um, you know, so uh, anyway, great piece. I was really pleased to see it. Um, anyway, read up, friends, it's a good one.

Lenny Murphy: Hats off, Jim.

Karen Lynch: Good job, yeah, good stuff.

Lenny Murphy: Um, this uh, now this next one, the Elaine Sedudo, this got really into the geeky data science weeds, right?

Karen Lynch: Yeah, I figured you'd like this one, right?

Lenny Murphy: But it was, it was. It was interesting. So, is another LinkedIn Elaine Zenudo synthetic data bias to Meng's big data paradox? Yeah. Warning that larger samples can increase confidence in a wrong answer. And I thought that was really, really interesting. Just because, but look, we tested against a thousand. That doesn't make it any more valid. So, well, I read that. You know what? Garbaging, garbage out, whether it's one or a million, it's still garbage. And the biases carry forward no matter what. And we can't analyze our way out of flawed inputs. So maybe be able to clean our way out of it if you know that it has to be cleaned. But that's, you know, this was in the world if you didn't know it. You're assuming that it's good quality data, but it has inherent biases in it. And just because you got more of it doesn't mean it makes it better. I just thought that was just an important point.

Karen Lynch: Yeah. Yeah, absolutely. Absolutely. I mean, you know, again, before we even bring up the last one, because we only have one more and we're doing like so well on time, Lenny, by the way. Keep, keep up these conversations and tag us also because sometimes we miss them. And like, I love this stuff. I love these thought-provoking conversations. And even if we don't chime in actively on them because we're having a busy kind of day or something like that, if Lenny and I can talk about them, we get really excited about that. So we like knowing about them. We love having conversations with you about them. When we're live, we'll chat with you about them. Like, these are the kind of conversations we live for here at Greenbook. So keep it up. We're so happy. Thank you. Thank you. Thank you. So, yeah. So. Steve Phillips, you know, again, again, what a week. So, you know, he's been writing this newsletter. I think it's, you know, AI squared. So it's, you know, a little bit of a little bit of a genetic thinking with, oh my gosh, I'm sorry, I don't have it opened right in front of me. But anyway, I'm forgetting what the AI square is, but it's AI and agile. But anyway, he's kind of posing the question about what he's talking about. He's labeling a kind of a system three, which is a genetic thinking. Which some people are arguing with him on, by the way. Like, some people are like, Yeah, no, like, let's not do that. But anyway, it's his prerogative. But it, but is it challenging the more traditional way brands grow, this Erringer Bass kind of thinking, which is kind of the way brands usually grow, which is how people, you know, kind of brands stay top of mind for people when they're making decisions based on what they've always done, which is like, oh, we stay memorable with our logos or with our colors or with our whatever else, advertising, et cetera, et cetera. And he's now challenging this with agentic thinking. And we've been talking about that, right? Like agentic thinking is changing the game. So it's just a newsletter article talking about the very thing that established marketing science is changing. And, you know, marketers need to be very aware of it. Traditional methods are not going to work in the new era of agentic decision making. It's true, whether it's called system three or not, it's a reality.

Lenny Murphy: What does brand mean to an agent?

Karen Lynch: It's the entire mechanism.

Lenny Murphy: We've talked about this for months and months. And this week, another big topic. Yes. Hat thought to Steve for coining the new system three. I hadn't heard that before. So hopefully that sticks and Steve will get the credit as the originator of that.

Karen Lynch: But the well, technically, technically, we I anyway, it's a long, long, long story about um when it's been used uh in the past, um, but okay, but I'm not gonna go there, I'm not gonna go there because I don't think it's the first time it's been used, but um, I just hadn't heard it before, yeah, yeah, but um, but but that's not the point. The point is, I don't want to get hung up on the semantics of it. What I want to talk about is what he's saying, yes, is. The concept of agentic thinking, being game-changing and changing the entirety of that marketing science, and what people have studied and learned and what's been taught in schools, is like ripping up the playbook because the rules of the game have changed. And I love that he's calling that out because he's a credible authority figure on it that people will listen to. And I'm hoping that academia is talking about that right now.

Lenny Murphy: I tell you, they are. There is a listserv in the industry called Research Wonks, been around forever, right? And it's more on kind of the ad, you know, kind of the ARF crowd, right? But I happen to be on it. The heavy-duty data science folks in advertising and marketing software.

Karen Lynch: Yeah. Yeah.

Lenny Murphy: Literally the last 24 hours. The biggest thread that I've seen happen on that listserv in a long time is how do you persuade an agent? And there's a lot of academics, a lot of heavy hitters in that. And a lot of people are having kind of fun with it, but there's also a whole lot of serious, like, yeah, it's the same principle. It is different. It is functionally different. It is.

Karen Lynch: Mechanistically, systemically different.

Lenny Murphy: And it cannot be ignored because all the other AI searches, blah, blah, right? I mean, it's becoming a big issue.

Karen Lynch: And yeah.

Lenny Murphy: All right.

Karen Lynch: So and on that note. Yeah.

Lenny Murphy: Yeah.

Lenny Murphy: It's a guy, it's just getting weirder and weirder.

Karen Lynch: It's true. It's true. It's a strange world, but I'm here for it. I'm here for it.

Lenny Murphy: Have you ever heard of Terrence McKenna? Terrence McKenna was a biologist who.

Karen Lynch: Anyway, he did a lot of weird stuff.

Lenny Murphy: He was very interested in mushrooms. Anyway, anyway, a theory, and it's a great quote. He's like, Things are just going to get weirder and weirder and weirder until people stop and say, What the hell is going on? And he was talking about just reality, but I, you know, I think it's kind of uh we see it playing out now, too.

Karen Lynch: That's so funny. That's so funny.

Lenny Murphy: And we're trying to help you figure out what the hell is going on.

Karen Lynch: Yeah, that's our job. That's our job. So, have a great weekend, everybody. Have a great week. We will see you. We will see you in September.

Lenny Murphy: We will. We will. The first Friday of September is going into a long weekend, but I don't think either one of us has any conflicts. So, we will be here.

Karen Lynch: We will be here. Absolutely. All right. See you next week, friends. Thanks, everybody. Bye. Bye.

Links from the episode:

Nvidia in Talks to Buy AI Startup Hugging Face, Reports Say 

NVIDIA reportedly discussed investing in Mercor at a $20B valuation 

Twin1 AI raised $20M in seed funding 

MERGE acquired Brado’s Digital and Insight practices 

NIQ and The OpenAI Deployment Company announced a collaboration to bring consumer intelligence into enterprise workflows 

Salesforce and Anthropic launched Claudeforce 

Zappi launched Synthetic Idea Pre-Screener 

System1 launched Test Your Ad Screen 

MFour launched DANI 

Phebi launched Real Time Response Capture 

The Model Context Protocol roadmap prioritizes agentic messaging 

Say it once: introducing Bot Preference Sync 

Finn Raben highlights synthetic data risk after the Median Strategies fiasco 

The Human Data Quality Paradox In The Backdrop Of Synthetic Data Use 

Elaine Zanutto connects synthetic data bias to Meng’s Big Data Paradox 

Steve Phillips asks whether System 3 challenges Ehrenberg-Bass thinking in consumer insights and brand growth