Separating AI Hype from Hope

Courtney: Great news. I just heard AI is going to solve all of humanity’s problems and usher in an unprecedented wave of achievement and quality of life. Uh. Hold on. This just in, I’ve also just heard that AI is going to destroy the human race via nuclear armageddon. Clearly, it can’t be both. For this episode, we’re gonna strip away all the AI hysteria on both sides as we separate the hype from the hope.

Hi, I’m Courtney Baker, and this is AI Knowhow from Knownwell, helping you reimagine your business in the AI era. As always, I’m joined by Knownwell, CEO David DeWolf, Chief Product Officer Mohan Rao, and Chief Strategy Officer Pete Buer. We’ll also have discussion with Geoff Livingston to get his independent, hype free view on the state of AI in business today.

Courtney: But first. Kind of regrettably, we need to revisit one of the predictions [00:01:00] David made during our Hot Takes episode in this segment called “Take a Lap, David.”

David, our team knows that you are highly competitive and

David: Oh no, I’m in trouble. What’d I do?

Courtney: we made some predictions on this show,

David: I know where you’re going

Courtney: Yes, as news has come out, Mohan and I have been. Regrettably saying, “Oh man, David was really right on this one.”

David: Yeah.

Courtney: Um, so I’ve invited you and Mohan back to revisit.

David: You had this brilliant idea about Hot Takes and I like, I actually, it was the one episode I like really prepared for. I’m like, I gotta think deeply, like what’s a real hot take? I even researched what the word hot take meant.

Courtney: yep.

David: Um, I was like, what does this mean?

And it, it’s something that most people would disagree with, right? That was the key [00:02:00] to it. And I predicted that it was the beginning of the end for the FAANG stocks, right? And, um, I really, I, I really felt like, yeah, Nvidia is kind of breaking up the crew. Um, but then I said, I’ve gotta get more specific.

Like that’s, that’s high level, that’s kind of easy. You’re picking a cluster of companies. And I said, you know what? It’s Amazon. Amazon’s gonna fall and nobody’s gonna believe it. Uh, but guess what? just released quarter two. Cloud provider market share. Guess who lost 2%? Market share. Pretty big loss. I think this is an early KPI.

I think I might have pegged this one.

Mohan: First of all, kudos to you, David. You called it right, the early indications, uh, seem like you’re on the right track. And Courtney, are we done with this episode?

Courtney: We are. That’s the end of it. I literally. Literally was just taking a screenshot of David because I thought, I dunno, that I’ve ever seen him [00:03:00] so genuinely happy as he is in this moment. Um, and listen, you know, David has a big platform. I’m sure his thoughts on the subject really, you know, he just influenced the outcome

here.

So.

David: was. I’m, I’m driving the markets right.

Courtney: That’s right. Uh, so yeah, I guess we have to give, uh, amends to David for amends. Is that that’s

not right.

David: I, no kudos, right? I don’t, I don’t need amends. Uh, hey, by the way, in that prediction, subtly, go back and listen. I also told you the reason why and I said how Microsoft and Google are out competing Amazon for this AI world. Guess what? Guess who picked up 2% Market share. Microsoft. Guess who picked up 1%?

Market share? Google. And if you go look at the why behind it, it’s all AI driven. So there you go.

Courtney: Man, that is, it’s pretty [00:04:00] impressive. Um, I would, again, we mentioned this in that Hot Takes episode, and I’m gonna double down on it. If you have such great power of making these things come to fruition, our team would all like for you to make a prediction about Facebook and see, oh, I’m sorry, meta. Um, and see if you can make that one come true as well.

But David. Great job, Mohan, you and I, I, I feel like some of ours are gonna come true. It’s just,

David: I bet you

Courtney: you know,

David: what yours were.

Courtney: do I, I, I don’t, I don’t remember Mohan, do you remember yours?

Mohan: Uh, no, I don’t actually. I’ve met so many, so who knows.

Courtney: All right. We couldn’t resist the opportunity to reconvene on this hot take and give you all an update. David, congratulations again.

David: Thanks.

Mohan: congrats. Great job.

Courtney: The AI world seems equally full of eternally [00:05:00] sunny optimist and doom and gloom pessimist. So how do we make sense of what we expect in the coming years? How do we separate the hype from the hope? I talked to David and Mohan about those topics in our recent AI Roundtable.

Is what I wanna talk about today because I think the media has really impacted how we think and perceive AI.

David: Mm-Hmm.

Courtney: so many different extremes. And today I kind of wanna break down. Where the hype is too big,

David: Okay.

Courtney: things that have been over exaggerated. And then let’s get really clear on where the hope is. Where are the things that actually we feel very strongly? We can kind of start looking towards and get excited about

Mohan: you know, the, the whole sort of framing here is very problematic from a media perspective. Uh, right. So this is a story in progress. And whenever a [00:06:00] story is in progress, uh, you’ve got both hope and hype. And obviously, uh, with the way you introduce the segment, there’s a lot of hype because that’s what sell, right.

So, so that’s my central framing for this. And the way I think of it is, um, for example, there is a lot of genuine uh, in the foundational models, uh, that are getting built. But the app innovation is completely. is a hype, right? Because you’re just projecting forward to what’s possible with these things.

And when I say app innovation, it’s what can you do with these things in a really meaningful manner? that’s not beyond the few use cases that we’ve talked about in this podcast before. And that is sort of fundamentally where there is this difference between hype and hope.

David: Mohan, the example that comes to my mind that I actually think has gone a little bit further in the cycle than the most of these kind of software apps, um, that have [00:07:00] been released is actually in the hardware space. Um, I think one of the areas where there’s great hope that is not overhyped because. They have largely failed on is the release of these AI devices, right? Um, the, the rabbit is an example, right? fanfare as they were released. Lots of buzz. Uh, people didn’t know. I think there was always questions about ’em, but they’ve really fallen flat. They’re not great. But I actually do have great hope that artificial intelligence is going to.

our usage patterns with devices and maybe able to present an ambient experience where we don’t have to walk around glued to our phone all the time. And I think if we can figure out what that device looks like, make it smaller and smaller, have it blend in, then maybe it can interject when we need it.

And hopefully we can build these way things in a way that [00:08:00] we’ve learned from history. That we shouldn’t embed all sorts of psychological ways for us to be emotionally attached and addicted to ’em. Right. And so I actually have a lot of hope there. I think it’s like almost pit of despair right now because they’ve fallen flat so much.

Um, but I, I think that’s an area where there’s gonna be some innovation that I, I’m actually really interested to see how it evolves.

Mohan: Yeah, I think, I think there’s a lot of hope there, but it sort of fits in the hype phase that’ll get converted to hope at some future point. Another good example is, um, you know, text-based generative AI is real. Uh, right. So we’ve all experienced it multi multimodal. Uh, is still in its, um, you know, early stages.

And what is even more in an early stage is how you would use this multimodal generative AI for healthcare and for creative and all of these things. So those things are coming, they’re not there yet. So everything around that people just project and there’s a lot of hype [00:09:00] around it, but you can see places where there’s genuine hope and people innovating with what we’ve got right now.

David: know what I love? I love the word you just used. How. How these language models are going to be used. I think that’s the crux of this. I think the, the hope is in the LLMs themselves. Wow. There seems to be breakthrough technology here. But the hype is in the applications of it. We’re in the process of dreaming and envisioning.

And when that happens, you have situations where those dreams go wild and crazy, which we need to pull us forward. Um, and you have these situations that totally fall flat and, um. Then you have to build up from. But right now I’d say it’s the application of the technology where the hype is, uh, if you want to abstract it, and more than just these specific examples.

Mohan: There’s so many examples of this we can give where there’s hype, right? So for example, [00:10:00] um, the, the, you know, in scientific research. truly something where AI has a huge role to play, but we don’t see the results of it yet. Um, and it’s coming and so people project it forward. Uh, I agree with you. It’s all in the application of these technologies where there is, um, a lot of future innovation that’s coming.

That rep that is represented as hype at the moment.

Courtney: Obviously we could have used the, I don’t know that the, the R one would’ve made the list, but if we would’ve recorded this episode, you know, a month ago, y’all could have said, Hey, y’all need to dial that back. They just obviously have fallen flat themselves and they don’t need you to call them out.

But is there anything else like that, that you’re like, yes, I think it’s gonna happen, but we have just been so much energy and excitement that it’s just over dialed in this moment.

David: The big one that I’m gonna pull it to is, is the social impact of this and going all the way to the elimination of work. [00:11:00] Like I, I do not think both from a technology perspective, we’re anywhere close to that, where that’s possible and from a humanity perspective, that we will ever get to a point where it’s actually good for people to sit around doing nothing.

Right. And I think. The amount of time that’s in the media because it’s salacious, um, because it stirs the pot, because it gets clicks, um, is just ridiculous. Um, I just don’t think it’s a relevant conversation right now.

Courtney: Are you saying no work at all or jobs being eliminated?

David: No, I mean, every time you have new technologies and innovation, there is a disruption of a job market. But I think for some reason, because AI is so scary and it reminds us of the big machine in the sky, um, we take that to be, um, you know, we’re gonna. I no longer need to work a 40 hour week. In fact, the vast majority of us are gonna be sitting around drinking lemonade at the pool all day.

You know, and, and, and I just, I, I get [00:12:00] perplexed where all of that comes

Courtney: Mm-Hmm.

Mohan: Completely, totally agree with that. Um, I think for business leaders, one, one other frame that they can think through this is not just be thinking about innovation, innovation, innovation, new things, right? So, but really think of it as a sustaining technology. Uh, right? So things that you already do, can you do better,

right?

Courtney: Hmm.

Mohan: So

that is a great place to

start.

Courtney: Yeah.

Mohan: As opposed to constantly be thinking about, Ooh, this is new technology and therefore I need to do new things. Uh, so it’s very much possible to be in sustaining mode and becoming more efficient and effective with, uh, this type of technology.

Courtney: Okay. I wanna flip the switch on both of you. Where are the areas that maybe you haven’t heard as much, it’s not getting the coverage that you think it should be? Uh, maybe it’s being under dialed.

David: I mean, I can’t help myself here, and probably all of our listeners have heard me say this in one form of function, but [00:13:00] I am continually flabbergasted. That we are not talking about how AI is going to transform the way we operate and orchestrate our business. I just, I, it, it blows my mind that we spend all this time talking about tools and productivity at a personal level, small personal level, you know, execution work.

Yet so much of what we do is actually how we all come together and work collaboratively together, and how you orchestrate that and operate that. And I’ll tell you, I am actively talking to VCs all the time. And nobody else is out there talking. They’re not seeing the stealth mode company, they’re not seeing others that are out there talking about true enterprise applications of this technology.

And I am so convinced that this is where it’s going and that’s where the major disruption is going to be. Um, I just think it’s under hyped. Um, I, but I think there’s a ton of hope there and it’s hope [00:14:00] that I’m excited about because I think it’s gonna actually. Elevate the human experience. It’s gonna help us be more engaged as employees.

It’s gonna allow us to do what we are uniquely qualified to do. It’s going to lower the bar of the unicorns that you need to hire because now they’ve got tools that can do the things they’re not great at. Um, there’s so much benefit to it, and I just don’t see enough people focused on it.

Mohan: You know, another area where, um. There is not as much hype, but there should be more, more and more hope, uh, is in the personalized experiences, uh, right along with what David said, which is elevating at a team level, at an organization level, the operations of the company being able to offer.

Hyper-personalized experiences. There’s some literature on it, but obviously AI has a huge role here in creating segments of one, right? So if you are using, uh, whatever application, it should just kind of cater to you. Uh, if you’re talking about if you’re a patient going to a doctor, you should have, you should, [00:15:00] the doctor should treat you as a segment of one.

That sort of hyper personalized, uh, it’s is massively big and, uh, you don’t see much, uh, discussion about it. There is some, but not a lot.

Courtney: That’s really helpful. I wanna go back, David, on something you said about really that, you know, orchestration, the operations of a business and how. We work together because again, yes, we are so focused on execution and me getting faster and better at what I individually am bringing to the table. Do you think that there is, the reason for that is there’s a disconnect between being able to take the problems that we have in a business.

We’ve kind of taken them for granted. We don’t understand how AI can actually. Completely solve that problem or change the way that we approach that problem. Does that resonate with you?

David: I, I mean, I think they’re much more complex problems. There’s no doubt about it, [00:16:00] which is why I think this is where the big impact comes from, right? When you’re talking about doing a singular job, figuring out how to do that job. Better is an easier conversation and it’s an easier strategy than thinking about how do multiple jobs come together and what are all the pain points in there.

I also love the word that you used of assumption. I think there are these underlying assumptions that we don’t even think to challenge, right? It’s easy to challenge, Hey, can I do this faster? But. Does the world fundamentally change? Does the way I collaborate with other people, does the way decisions get made fundamentally change?

I, I love the example Mohan has used a couple times. I’ve heard him use it, uh, about like right now. You, you have this org chart, has departments and silos, and when we host a meeting, we just have to be inefficient [00:17:00] because nobody has the single and all of the information they need to make the decision.

So you invite finance and you invite contracts, and you invite engineering and you invite sales because there’s a piece that they may need to opine on, but people sit around, well, can we just totally change the dynamic where no longer. Do you have to have everybody in every meeting and where we may be able to live in a world that doesn’t have a hierarchy as the organization structure of an organization where meetings aren’t the end all be all, and only way you can debate a subject like underlying assumptions like meetings and org charts is what we’re talking about.

Mohan: What, what I love about this example is it’s a specific use case that you take because Courtney, the question that you asked is a very complex question, right? So. You know, there’s so many dimensions to it. Uh, individually, generally we are efficient. Collectively we are less efficient, uh, right. So that’s true of a football team.

That’s true of any business organization. [00:18:00] Uh, right. So there is that problem here to solve, but the way to pick at that problem is to go after use cases. And the one that David mentioned about knowledge management is. A huge opportunity for, uh, for AI to solve and for democratize the knowledge. So anybody, if I want to know when a client contract is gonna come to an end, I should just be able to ask and no, not have to go to somebody in contracts or legal, and they look up some PDF, that’s somewhere, and then tell me the answer.

Courtney: Yeah, and I would be amiss if we didn’t also talk a little bit about Knownwell and what, how we’re changing how you think about client intelligence and how that changes the operations of your business.

Mohan: what we are building is a commercial intelligence platform. can get true intelligence only in the act of the commerce, meaning goods and services being exchanged and, and the [00:19:00] intelligence you can mine from those transactions.

Or pure gold compared to anything else that you can get at. So the idea of the big concept that we are working on is how do you get this commercial intelligence? How do you represent client knowledge in the platform? How can you, uh, figure out the, um, the root causes of issues? Where are the growth opportunities?

These are the big problems we are solving.

Courtney: David Mohan, I am so proud of you. Thank you, uh, for your work in taking on the media industry today. Hopefully for our audience, uh. Found this helpful as we continue to, you know, we’re over a year into this, uh, but I still think we continually have to figure out where are the things that we’re, we’ve overhyped and what are the things that we actually are under hyping?

David: And if anybody wants to Overhype Knownwell, feel free.

Courtney: Good point. I like that. Uh, the CMO should [00:20:00] have, should have beat you to that one. Uh.

Mohan: but as the product guy, I’ll tell you that there is real hope as well with

David: The, the,

Courtney: Nice. Thank you

David: our hope is longer than our hype right now. We need to hype it up.

Courtney: David Mohan. Thank you.

This Thursday, May 30th, I wanna invite you to a special discussion with our CEO David DeWolf and CEO of Teer, Chris Barbin. They’re gonna be hosting a round table with leaders of professional service companies to discuss. Client retention in the AI era. If you are one of those people, a leader at a professional service company, I invite you to join us.

You can sign up today at Knownwell dot com slash roundtable.

Geoff Livingston is the founder of [00:21:00] Cognitive Path and host of the No-Brainer podcast that talks about marketing and AI, and I recently was a guest on his show. It was an absolute blast. Geoff sat down with Pete Buer recently to talk about where he sees AI having the biggest impact in the years to come.

Pete: So Geoff, great to have you on the podcast. Welcome.

Geoff: Thank you, Pete. It’s really a pleasure to be here.

Pete: If we could frame things up for our listeners please, to give some context before we get into the conversation. A little bit of background on Cognitive Path and your role.

Geoff: Yeah, I’m one of the founders and principal analysts where we really are as an advisory that helps enterprises and associations incorporate AI into their, uh, business processes with a strong focus on marketing.

Pete: the description on your LinkedIn profile, which reads hype free views on AI and digital media. Let’s start with something, uh, [00:22:00] controversial or, or let, let’s start with some edge. Where are we hyped out of our brains on ai?

Geoff: Geez. You know, I mean, I think we kind of know the generative AI thing is a little bit, uh, sharky right now in the sense it’s jump the shark

I really feel like what we have, and you and I are both at startups and we may be just as guilty as, as some of the players in the business, but we have the usual kind of Silicon Valley.

This is gonna be a panacea to save the world type of thing. Or, and in some cases, some doom saying, and in reality, generative hasn’t really met the. Point of truth where it really makes a huge impact for most enterprises. Some are using it well, but generally most are not. And so I think that’s really the hype wave, and I think it’s led by open AI first and foremost.

Pete: On the doom to panacea spectrum. Where are you then, Geoff?

Geoff: I, I’m probably five. I would prefer neither.

Pete: Yeah. Right. Okay. So it, it is what we make of it, I suppose.

Geoff: Yeah,

Pete: [00:23:00] so we’ve, we’ve talked hype. How about, how about hope?

Like, so generative AI hasn’t yet met its full potential. which areas do you think ultimately it will? I.

Geoff: So this is gonna sound crazy and I’d be interested to hear your take on it, but I personally feel like the real issue with this in my mind, at least from an enterprise standpoint, is the failure to blend generative with ML and traditional analytics, and even automation in some cases, because, uh, generative really works well when it’s fueled by strong.

Customized analytics, and I know this gets to some of your value proposition as well, but without that deep insight that’s gonna make that worthwhile to deploy some sort of generative text, generative image, something that’s gonna personalize it or make it unique to that particular type of account, if you would

Pete: Mm-Hmm.

Geoff: it, it really kind of struggles and you really have this kind of.

Wonder, wonderful technology that [00:24:00] could theoretically write well or do something unique with a video, but may not be valuable.

Pete: Interesting. So I, in, in my mind, I’ve, I’ve split the world up into two parts, maybe in the same way you just waved at, there’s kind of generative AI for the masses there’s proprietary value prop driven, either blended ai or, or, or, um, bespoke for u unique value propositions for, for customers.

Geoff: Yeah.

Pete: Um, and, and I, I can see how the second.

We’ll have ROI for the first. What will it take to get generative AI for the masses to start driving a return?

Geoff: Uh, I think it, I think it’s product marketing and product development. To me, I, I think what we’ve really seen. Happen is that we’ve got some really great algorithms that could do incredible things. And if you think about the way a traditional data science [00:25:00] team would use one of these algorithms, they would’ve taken it to a specific problem and then train it or incorporate it into a larger model and train it to address that problem. And what we have. Marketed is fantastic. Chat technologies, fantastic video technologies, fantastic image technologies that are really, really powerful in some ways looking to find a, a solution or a problem. And I, I look at chat, GPT, like, did you see what happened with 4.5 turbo? Again, open AI is the easiest one to pick holes at because they’re the most visible and they’re the 800 pound giant.

Of course, without them we wouldn’t be where we are, but

Pete: Yep.

Geoff: turbo. They just released 4 5, 4 5 or whatever it is,

Pete: Right.

Geoff: four O turbo. And uh, they made it less chatty. Right? A little less verbose. And it just goes to show you like it was meant to be this incredible talking technology that anybody can use to write and.

And the [00:26:00] reality is, is that people found it to be a little bit windy and not very good. And uh, that’s common feedback you hear amongst the marketing community. In fact, I think writing amongst marketers is the second or third most popular use case, not number one.

Pete: So, um, is, is generative AI in the long run gonna be just, um, basically a, a tax, uh, on the business? It’s a, it’s a, uh, a subscription that you pay for and, uh, you don’t get returned from.

Geoff: I, I don’t, I don’t know. It does feel like it’s getting incorporated into everything, right? Like, I think copilot office, so as a productivity tool, probably so, right where it’s a $20 a month tax.

Pete: Yep.

Geoff: But from a functional standpoint, as far as how it can really serve an enterprise and achieving its goals, I think like the rubber really hits the road when enterprises start training it and using it within their own private instances and take it to a level beyond that [00:27:00] productivity tool.

What do you think?

Pete: what I, I think ROI doesn’t happen with the, with the idea or the application. It’s in the transformation and change management and re-skilling of the business and the driving to ROI. And so I think it’s great to put tools in the hands of, of folks, but until we. Reassess org structure, reassess, uh, role profiles, reassess human capital requirements in the business.

We’re not gonna drive change. We’re just gonna add a tax, an equipment tax to the business.

Geoff: Right. And when you do that work, you’re identifying the problems where this can actually be of service, right? Like, Hey, I have to fill out the this form a hundred times every month. Maybe we can use this for that. Can we build some custom templates? How can we do something with this?

Pete: Yeah. And, and, and drive it through, right? Communicate it, get paid for the innovation with your employees, reassess your, your talent load. Yeah. To totally with you.

Geoff: [00:28:00] Yep.

Pete: Okay. So, um, you mentioned marketing and I know that’s a particular angle.

Geoff: love marketing.

Pete: Uh, for you,

Geoff: I,

Pete: where are you seeing the most cool stuff going on in marketing?

Geoff: To me, I think the real value is personalization. And uh, of course we’re not seeing a lot of that right now, but one of the things that’s happening, and there’s like so many different aspects of the technology industry that are happening right now, but with GDP are, and now with the AI Act coming out of, uh, Europe, I forget the formal name of it.

Um, I mean, what you’re looking at is a lot of. Interesting behavior patterns that are getting forced on the digital media companies to protect users. And so third party data, it’s gone. Maybe first party data may be next. Who knows? Right. And so companies are being really forced to maintain their data in a way that they hadn’t done so before.

And I think every enterprise, [00:29:00] large and small is gonna be in a place in the next two years where they have a kind of a come to. You know what moment and really have to clean this up and to do that, then they’re gonna have to deploy ML to really, uh, get a better job done with their marketing so that they can speak to specific segments, speak to specific people where possible, and that’s where generative and uh, other technologies are gonna really come in and help organizations market better.

I think.

Pete: And, and are there. Either applications and development that you’ve seen, or is there stuff out in the world that

Geoff: Oh yeah,

Pete: listeners could, could, could look up and learn about that are super cool?

Geoff: yeah, yeah. There’s quite a few CDPs just about every CDP out there right now is saying that they’re doing this. Um. I think what’s interesting is to see what’s happening kind of in the, uh, Salesforce universe, right, within Einstein and how they’re basically letting their partners really tool this thing [00:30:00] up.

One thing to really look out for is Google and HubSpot, ’cause they’re about to get, uh, merged or they’re looking at merging together and they’re in talks. And that would be a massive change, I think. Uh. The other thing to look at too is what are Accenture and the big consultancies doing with this right now, and how are they approaching it with their customers?

’cause this is really classic management consulting, implementation material.

Pete: And, and if I’m a, if I’m a CEO listening to the podcast wondering how these new combinations of companies coming together could. Create a value proposition for me that’s different and worth snooping out. What would I expect to come out of those alliances or partnerships?

Geoff: Boy, I would be really suspicious to be candid. Uh, I think, uh, what they, what I see from most of the vendors is they don’t really know.

Pete: Yeah.

Geoff: so like, I saw Adobe present last week and they were talking about their marketing cloud and how they’re going [00:31:00] towards personalization with all the creative firefly tools and pairing it with, I mean, what used to be Marketo, right?

And, uh, but when you got to brass tax, uh, all you heards, well, 30% productivity increase, 40% productivity increase. Sounds good, but how.

Pete: Well. Alright, so Geoff, um, the way I, find I get my best learning about AI is talking to smart people and today was one of those days. So thank you so much for, uh, sharing your experience and your insights and for joining us, uh, on the podcast.

Geoff: Pete, you’re a scholar and gentleman, and uh, thank you for having me. Really appreciate it.

Courtney: Thanks as always for listening and watching. You’ve heard me ask before, but it’s no different at the end of this episode. It would be really helpful if you would leave us a rating or a view. It is the one most important thing in helping more people find this show, so we don’t make you listen to any ads, but we would like for you to give us a review.

At the end of every episode, we like to [00:32:00] ask one of the large language models what their opinion is on the topic at hand. So, hey Claude, welcome back to the show. This episode we’re talking about AI hype versus AI Hope. What do you think?

And now you’re in the know. Thanks as always for listening. We’ll see you next week with more headlines, AI Roundtable discussions and interviews with AI experts