AI and CX: The Shift Toward Smarter, Seamless Service

Hey, everybody. Welcome back to an episode of the Modern Customer podcast. I'm your host, Blake Morgan. Today my guest is Zach Cass. Zach is a futurist. He is the founder of ZK AI Advisory and the former head of GoToMarket for OpenAI, where he built its sales partnerships and solutions teams. His work as an executive in residence at the University of Virginia focuses on the socioeconomic impact of the AI revolution. And his advisory firm helps Fortune 1000 companies and governments effectively adapt to the rapidly changing environment. Zach's mission is to demystify AI and help the world build a more beautiful future. Today we are talking about AI and all things customer experience. We're talking about his predictions for how AI will change customer experience overall, the software market and so much more. Please enjoy Zach Cast. Zach, welcome to the Modern Customer podcast. It is so nice to see you this morning.

Thanks for having me. It's great to be here.

You're about one hour north of me in Santa Barbara, so maybe next time we can, like somehow record this in a studio. But it's good to see you. I've definitely noticed you on the speaking circuit talking a lot about AI. In my audience. We were just chatting. Are largely customer service professionals. Contact center leaders, CXOs that run customer experience. I think we should just kick off. I'd really be curious to know how you actually got involved with OpenAI. And I know you started doing AI stuff really early, but can you just talk a bit about your career?

So I graduated Berkeley in 2009 having studied history and computer science and then actually tried to play make a go of it, playing volleyball, which didn't work. But it was the global financial crisis and so there just weren't many jobs available. Came back to San Francisco, got a job at a company called Crowdflower, which then became known as Figure 8 and then sold to Appen and is now part of the Appen business. It was the first machine learning data labeling service at computer scale. And so it was a quality layer on top of Amazon's Mechanical Turk, which OGS will remember, was this service that Amazon built basically to reward shoppers for doing small, very rote tasks for them in exchange for Amazon points. And the Amazon team used it to improve their search algorithms and a lot of their product descriptions. So we were really early in that world and that gave me this incredible insight and launchpad into machine learning and AI that led me to another company also run by two Stanford researchers at the time. Early, early in neural nets, a company called Lilt, which was doing machine translation Powered by large language models. And then that led me to OpenAI.

Mm, got it. I know our audience are gonna wanna ask this, they're gonna be interested in this, and I'm interested in this too. What was it like working at OpenAI? What was the culture there?

I mean, I would describe it as one of the honors and joys of a lifetime, I think. Yeah. There, you know, I got, I got called when I left by a lot of journalists looking for dirt. Now it's like, look, I have bad news. I have very disappointing news for you. I have only raving positive things to say about the experience because it truly is a once in a generation company that proved a lot of things. But one of the most amazing things about working in OpenAI, in addition to the talent concentration, is the optimism and positivity. It is a company that radiates positivity

and

a culture that has, I think, done as much for the. I mean, obviously some people may disagree with me, but AS has done as much for the sort of understanding, the global understanding of AI as it has for actually AI research. I mean, and so one of the incredible things about being at OpenAI is not only were we building AI research, we were shipping AI products, which is just not something that Google and Meta and others were doing even though they had the ability to do so. So I would describe it as an amazing place that gave, that sparked this incredible fire that now the world is appreciating as this AI gold rush, but also a place where I made a bunch of amazing friends and learned a whole lot and just will never forget.

So I spent some time down the rabbit hole of your talks and watching your keynotes, and you're clearly an optimist. You're excited about the potential for AI to make people's lives easier and better. You've said it's going to reduce suffering, increase joy, solve fusion, cure cancer, nanotech, quantum computing, diabetes, cure diabetes, like a lot of very positive things. This is a customer service podcast, so we'll kind of stick, stay in our lane. But I mean, why do you think that. That AI is good overall for customer service and experience as an industry?

So I guess maybe before we talk specifically about customer service, and we can talk plenty about it, I have a lot of thoughts. But what's important to acknowledge, anytime I spend any time, I'm engaged with someone who doesn't believe that the world is getting better. So either they see themselves, some of these people will describe themselves honestly as pessimists or cynics. And some of These people will describe themselves as realists. They say, look, I just see it as it is. And I say, that's interesting. Tell me more about how you see things today versus how you would have seen them 100 years ago or a thousand years ago or 10,000 years ago. Because from my vantage point, having studied history and having also studied technology, it occurs to me that the world gets better all the time. And that in fact, recently the world has started getting better at a much faster rate. And that we now live in a world that is actually that, where the standards are so high that any short term regression in any of these things causes like incredible outcry. And I'm not arguing that the world is a perfect place by any means. In fact, I would argue that the world has so much more to improve upon. I would also remind everyone we crawled out of caves at one point. I mean, depending on exactly what you believe, humans definitely came from a far more rudimentary place at one point. And what's clear is, is that we basically measure progress on how free are people and how inexpensive are goods and services. And we are getting more free all the time. People on Earth and goods and services are getting less expensive. And these things are actually somewhat correlated, it turns out, because it's very hard to run a very unfavorable government when the economies get better, because people start looking around and saying, hey, I don't want to live in a prison when I can enjoy so many luxuries. So my view on the world is simply that it gets better all the time. And it's very. The onus is on someone who believes it's getting worse to actually argue. And I also think it's somewhat egotistical to say we are the peak of civilization. I take this look that in fact we're very, very early in our species journey and that there's a lot more to go. So that's just for the listener's edification. Just to admit my priors, this is what I believe. And so anyone sitting there saying, well, this guy's, this guy's crazy or naive again, my challenge is simply, you need to prove that the world is getting worse. And that requires actually remembering that 20% of the world's poverty population, 20% of the population that was impoverished, has been taken out of poverty over the last 30 years. And you really have to start to question what you believe is great versus what you've come to expect as sort of standard.

I agree with you. I will say, okay, okay, great. I'm an optimist. Because every time I pick up my phone and I order 20 things from Amazon, I remember, like, the schlep of my parents having to go to the local store. Like, I wouldn't be able to be a working mom and a woman in business, really, if it weren't for technology, smartphones and social media. So I. I'm with you. I got you. I agree.

All right, so. So. So there are my priors. I just want to. I just want to spell that out so that no one. I'm okay being accused of naivete, but I do think people who believe the world is getting worse really need to. Really need to examine how much they have to prove in order to. To. To make that case now with customer service. Yesterday I had this incredible experience that I was really excited to talk to you about, wherein I ended up playing literal telephone with a very large hotel brand that will remain unnamed, who I am actually very loyal to. And it ended up. It ended up costing me about an hour and a half. And it was work that basically I alone could do because it required navigating a bunch of systems, some of which my assistants didn't have access to, and in the end, had to actually help some of these employees make sense of a. Of a system that they themselves didn't understand.

Okay?

And it got elevated to multiple people. And the whole crux of this was consolidating two. Two loyalty profiles. One of them had been sort of established rogue at one point, and the other, you know, was my primary one. And this one had been accruing some night stays, and it sort of broke the system when I had to make these things work. And it required that I actually call multiple hotels to get receipts and then consolidate these things. So I was left sitting there saying, okay, who wins? Because this, like, who are the. Who wants this to stay the same? So I was trying to, like, make, you know, you know, you can, you know, follow the money, right? It was like, you know, corporates are greedy, right? That sort of. It's consumers priors these days. And I'm like, this is not a result of greed. There is truly no one's interest is actually in the messiness of this process. Not mine, not the hotels, not the people running this function, not the people delivering the function. And so then I sort of sit there and say, okay, why did this happen? Well, obviously it happened because we built an Internet that allowed us to connect a lot of information, but it also doesn't, like, perfectly do so. And so you end up having these magical moments piecemeal Together by these very unmagical awful moments, right, where I'm like literally getting a physical receipt and sending it to some, you know, un, unnamed email address. And so I look at this and I say, look, it's 2025. And we talk about all the incredible things that, that we are going to have soon and the number of experiences that exist like this. And this is an advanced hotel company. This is a very advanced business. And I'm like, no one wants this, no one wants this. And this is clearly the stuff that's about to be solved by AI, by agentic AI, where all of this stuff starts to become automated in a background. And then that leads to the obvious question, which is what is the future of customer service? And what does it actually mean to provide great customer service? And in a world where everyone wants to differentiate, it occurs to me that actually figuring out how to start doing more proactive upstream work becomes the future of customer service, where the back end actually starts to feel commoditized. And so my theory on why it all gets so much better is that it starts to automate a lot of the work that accrues a really negative experience to the actual user. No matter how wealthy they are, no matter how much status they have. It's just stuff that has to get done. And then it allows these companies who pride themselves in great customer service to start seeing customer service as a true value add, as actually a differentiating experience wherein they can start doing really creative, interesting things because the stuff on the back end isn't so. Isn't so rote and quagmired. And this was sort of my take from yesterday and I guess I'll sort of start and leave that there.

Yeah, I like that. Because it's like so stupid that most of us that study customer experience are just scratching our heads going like, who wins when you treat customers like when you're so inefficient that literally why? Why? And it just seems to boil down to limited thinking, like short term thinking. That is so rampant still. I mean, that's my answer.

Do you think it's limited thinking? Is that. I mean, you know this way better than I do. Like, is this hotel broken in this series of processes because. Because someone hasn't thought about it enough

for customer experience to work. The whole way the company is run has to change. Like the way they measure success, like how siloed these companies are, how anything with the word customer in it is seen as a cost center. The way technology purchases are made, the way success is measured. If I haven't said that already. And it reminds me of the CTO of Sephora who did their digital transformation, Ollie Bauhauch, who before doing any technology implementation he had to knock all the walls down of the whole company, metaphorically because the E commerce competed with the stores and the app and all the divisions were competing for money and we had to knock them all down to create new KPIs to, to where the teams wanted to work together and create a seamless zero friction view around the customer. But I mean it boils down to how these hotels, they probably compete and they don't share data across all the hotels. And there's all these, these companies are so big, it's probably a hotel brand with like 50 different hotel brands within it and you have different accounts for each hotel.

That is right. But, but then I did some digging and discovered that these hotel, the hotels that I was staying at were owned by the brand itself.

Yeah.

So then the question is why? Why? Yes. Maybe the issue here is simply the, the company needs to redesign its incentives so that the individual hotel managers don't feel like they're competing with each other and therefore are more incentivized to aggregate data. But if that's actually their problem, then you're right. Like the, the limited thinking. I mean that's such a incredible misaligned incentive. I don't even know where to begin. My sense, and again, maybe this is wrong, is that this was simply a result of people not having actually invested the time to solve this problem because it was deemed either an edge case or a problem that just couldn't be solved. Simply that it's like one of these things that we just, that we simply live with as like technical debt which by the way exists in lots of software companies. I mean so many software companies in the world have bugs that are known bugs or terrible product features that the head of product has simply said it's just not important enough. We know that people hate it and we have too many other things to do. Which is sort of how I see a lot of customer experience issues. But again, maybe I'm wrong.

Yeah. And that's why the more customer centric companies are often run by founder CEOs. A lot of them are probably companies you've worked with or been customers of digitally native. Because the founder CEOs can be like Steve Jobs and just tell everyone, like tell the board like, sorry, this is what we're doing because it's the right thing to do instead of these other CEOs that come in and they make decisions based on how they can. I mean, it sounds terrible, but I mean, they're thinking about their bonus. They're thinking about just looking good for the next few quarters until they take their next CEO job. And I do think that the leadership, it's a big piece of it.

I'm sure you're right, but I would still point to, I mean, yeah, maybe Travis Kalanick is an interesting take, maybe he's not a customer experience. Brian Chesky is a great example, right? Airbnb is one of these classic founder CEOs customer experience first. And Airbnb had so many customer experience issues by nature of the fact that what it was doing was so complicated. I don't think, you know, Airbnb is one of these fascinating companies to me that I, I really just think it's not nearly enough credit for. I don't like Airbnb for the most part. I don't use it. I much prefer the reliability. And when I use it, I use Airbnb Lux because they control a bit more for it. Okay, but, but the amount of work that is required to organize a marketplace that big and make sure that the number of the multiple constituents, it's a multi sided marketplace. The number of constituents involved are happy, and not just happy, but thrilled. Because the expectation when you rent a car is very low. You're like, does it have gas? Are all four tires good? Does it smell like smoke? No. Okay, fine. It's acceptable. When you stay in a home, your expectation is, you're like, this is. Know this. I'm comparing this to hotel and what they have to do to coordinate all that is incredible. And as a result, a lot of things did not work very well because there were just so many things that, that people were judging, right. That that needed to work. Well, they messed up a lot of, you know, they've just messed up a lot of stuff over the years. And my take has always been because they didn't have the tools to actually make everything magical. And I think in Brian's case, he. And one of the great reasons to bet on Airbnb now is because of AI, it can start solving a bunch of much more complex issues that Brian would love to solve, but he simply cannot. Because the number of moles that you have to whack in a game where you are serving multiple sides of a marketplace with very high standards is just crazy. And so my argument would be that in Brian's case, he's a founder CEO who really wants to deliver a magical customer experience, who has simply struggled to do so consistently because it's just been this hard and now he can. And my, my guess is in the case of this hotel company, maybe not, but my sense is maybe it's the same. Now the question is, are they going to adopt it as fast? And this, this leads us to the actual interesting where the rubber meets the road, which is now everyone's going to have the tools available to them. Who's actually going to deliver? Right. Who's, who's actually going to take advantage of this technology and start to set themselves apart from the rest AI, customer experience.

The whole software market is in such a state of flux right now and maybe even panic. Where all these companies now have to be AI companies, some of them are even exaggerating their features. It's just such an interesting time to be studying customer experience at the same time. Yeah. Like some of these blue chip brands, like these old school brands. Yeah. The experience delivered just still falls so short. And then you being this technologist, you're going, why, why? Any, any other thoughts about what is going to happen to the technology world as a result of, of OpenAI and everything it has done to disrupt technology?

Well, that's a, I guess a big question. Yes. My expectation is that, as you just said, I think a lot more, I think software companies will start to label themselves as AI companies and. And so will probably many businesses. I also think there will be a fundamental, I think, you know, one of the questions is, does the next great, you know, does the next great law firm exist today? Does the next great hospital group exist today? Does the next great insurance provider? I mean, now you have to start asking, with a paradigm shift this big, will industries be reinvented from the inside out? So will we see hospitality brands adopt this technology and redefine what hospitality and customer experience should be? Or will you see an upstart like Uber take advantage of mobile the way it did and do so with AI? Right. So whereas the next great taxi company did not exist in 2005 or 6, the next great hospitality brand did not exist in 2005 or 6. It took this mobile revolution to introduce the world to Uber and to Airbnb. The question now, I think broadly should, should actually be asked of the entire world and every major industry really, with the exception of things like rare earths and mining. Probably these companies are going to. They're like cockroaches. If you own the earth, you just live on past everyone. But I think many businesses are going to discover that their moat is actually much smaller than they realized when the cost of creating value starts to shrink and so panic is going to give way to, I think this renewed, whatever you're observing to this renewed sense of optimism and excitement, people are going to say, yeah, well why does this have to be broken? And more importantly, is this company even the right one to take the industry forward? Or, or are we going to see this reinvention of, of these, of these companies? And, and as a reminder to everyone, The S&P 500 turned over a lot in the last 30 years because of the Internet and then mobile. Why is it, you know, why are all the blue chip stocks guaranteed to be blue chip stocks forever? I mean there is no guarantee, right? I mean there's, there's just no reason that we won't see a bunch of new entrants into, into the, into the global economy because of what AI presents.

But let's just for a minute, let's just look at some of the most hated industries in the world.

So government from a customer experience standpoint.

Yes. Health care, telecom, insurance. I mean, let's look at insurance. I've been sort of airlines. Oh. And yeah, and airlines. I mean let's say like, do you think that like let's say insurance. I mean we both live in California. You and I live close by. My house almost burned down in January. We still have home insurance.

Wow. Where do you live?

Calabasas. Yeah. So I guess the question is will these companies use AI for good to, to make customers lives easier and better or will they just like you could see insurance companies like you, they're using drones to minimize their risk and like fire customers to see like how much brush is in their backyard. Or will they use technologies to figure out ways that are like solutions that are win win. Like how can we maximize our profits while also helping people that are in need? What do you think?

Well, here is where I'll have to show some political colors and maybe I'll lose some people, maybe I'll win some followers. But if you want to know the risk of a company's ability to operate in a predatory way, you only need to ask how much lobbying exists on behalf of that industry and therefore how much regulation is actually designed around the industry. And the part of the reason that insurance is so broken is because it is so regulated. There are a few companies in the world, in the United States that have basically created this oligopoly and convinced us all that insurance will only work if a few entrants can exist. And the amount of lobbying dollars spent in Healthcare insurance basically dwarfs that of most other industries. And so energy is sort of up there, defense is obviously up there. And so what you discover is you start to, if you want to red pill yourself a little or if you want to sort of explore this, is you really only need to go and figure out where is the money being spent by lobbyists in D.C. and it's a bipartisan effort. Both sides of the aisle are equally susceptible to lobbyists. If you actually peel back the opportunity for new entrants and for people to change and revolutionize industries, then you start to answer the question, how much will this technology meaningfully impact and how much will the profits and the value be distributed to the marketplace? And we saw this with. We started to see this with industries like Airbnb. We started to see this with, with or with hospitality, with. With. With. With airlines, et cetera, where we started to see, like, companies show up and sort of reinvent the industry and set new expectations for people. The question is, is, is that going to be a lot? Are we going to allow that to happen in things like insurance and health care? And my answer ultimately is yes. And I think what's going to happen is we're going to see the privatization of, by the way, education, to me, is where this is going to start. Because education is fairly deregulated. The number of parents that are already moving their kids, for example, into micro and home schools is growing about 8% year over year in the United States.

My friends talk about an AI school, by the way. They want to pull their kids out of public school and put them in these, like, AI programs where the few hours a day they're training with AI and then the other hours like they're doing.

Is this, is it Alpha School?

I don't remember the name. It's in Texas right now. And Alpha School D families. Because a lot of us don't really like the.

It's Alpha School curriculum.

So.

So Mackenzie Price is, Is a genius, and what she is building is incredible. They're opening one in Santa Barbara. I, I'm, I'm obviously, you know, I'm particular to Al. I don't. We're, We're. We're. We just got married. But the Alpha School model. And again, I don't want to. We're gonna. I don't want to leave people. I don't want to wander too far off the reservation here. But, but, but what, what people need to know. Alpha School is, is the future of education. Now, Alpha School itself may not be the, May not be the the only winner, and it may not even be the big winner. But what it is proving is that the tools required to educate children scholastically in a much more efficient way now exist. AI powered private education. AI powered one on one tutoring specifically is going to be the future wherein we can do much more interesting scholastic learning in a much smaller window and then let the child explore the world in a social life skill sense in a much larger window. And my wife's a first grade Waldorf teacher, so I believe very much, I believe very much in alternative education as a means to sort of raise the whole child, especially in a world where intelligence becomes very inexpensive thanks to AI. What you know, actually is far less important than how you make someone feel and how you move through the world. This is something I believe very much.

Yeah.

So what's amazing about education is then the question is, okay, are the public schools going to adopt what McKinsey has branded as two hour learning, or are the, or are we going to reinvent the system? And what McKenzie did and what two hour learning and Alpha School did is they said, look, instead of selling software to schools, we're going to build our own school. And my theory is that this is now going to happen in every major industry. What we are going to see is people say, you know what, I've had enough of this industry entirely. The answer is not selling software into the industry to make it more efficient. The answer is redesigning the industry with the tools that we now have available to us. And for those who are so cynical that they cannot imagine insurance getting better, I will remind you that the world gets better. And it is truly so abysmal, so abhorrent right now that it can only get better at this point. And it will. And I think it's going to get better for two reasons. One, what we're seeing now with education is that in a world where parents can take their kids out of public school and put them in private school, people are building much better private alternatives and giving them the opportunity to do that. And then the more parents that do that actually, the lower the cost of these charter and private schools because you just have economies of scale. The other thing that's incredible and happening in real time is parents who can't afford Alpha School are actually realizing that the tools that Alpha School uses to our learning are available to them. And so the open sourcing of these one on one private tutors is now so available to parents that they can say, you know what, we can actually design our own curriculum for our kid. And we don't have to put, they don't have to sit in a classroom with 40 students. The same thing is going to start happening in all these industries because the cost to deliver the value in that industry is going to come down. And primary health care, family medicine and pediatrics is next. I just assure people that there will be many interesting alternatives to their current primary care setup. Today. In a world where GPT4O is already better pathologically at diagnosing 180 skin conditions than a dermatologist, where it is better at basically every radiology scan, except for the chest scan, than a human radiologist, we are starting to basically make the cost of preventative and primary health care so inexpensive that companies are going to show up and say, hey, we will deliver you this service outside of network, no insurance required. And it basically takes then all of this really, really important, low cost stuff and gives it to families at a much more accessible price. And you can start to imagine local businesses starting up and saying, hey, we will protect your, we will guarantee your house with a drone. I mean everyone, everyone bashed Rick Caruso for protecting his investment when the Palisades fire hit. And I get it. And also he knew he laid flame retardant. He knew what the risks were and was prepared. And people are like, oh, it's so dystopian to describe private services. And I'm like, is it actually, is it actually that crazy to imagine a world where the government becomes such an inefficient spender of capital that we actually see this incredible new influx of, of services designed to be much more competitive in the market in a world where the California government doesn't have to compete with anyone. So they don't. Right. I would love to see private companies start to compete with Newsom on the services that he offers. Anyway, we've gone really far at this point. But my point to most people is if you hate an industry, it's not long before someone very pioneering and smart realizes that that industry can be reinvented because of the tools now available to us. And the, and the question on how predatory the process is going to be lies in how much does the government get to control the regulatory capture Insurance companies profit because the government has allowed them to profit so much and has created laws specifically to let them do so. And if we actually create a competitive marketplace, what would, like, what would the outcome look like? And I think we're going to find out over the next 20, 30 years.

Wow. We could honestly talk all day. Um, this has been really interesting. And I hate to say we have to get into the rapid fire before I let you go number one. What's the most important part of your morning routine?

Waking up and kissing my wife.

Aw. What do you do to relax at the end of a hectic day?

We have a sauna and a cold plunge on the patio. Yeah. So, yeah, it's. It's a real. It's a real Joe Rogan answer. Or we. We live in Santa Barbara, and so we. And we'll often take a walk down State Street.

Oh, awesome. What show are you currently streaming?

I have not watched a show. I've not watched a season of a show, start to finish, since Game of Thrones because I just don't watch a lot of tv. Always sunny in Philadelphia is my guilty pleasure, and when I must watch a screen, I watch it. But we did just start White Lotus Season 3, so I'm hoping it gets interesting soon because it is a slow burn so far.

I know it's sort of. It's not as good as the first two seasons. I don't know what happened.

It's hard. Those are good shows.

I know. It's a high bar. What's your idea of perfect happiness?

Easy outside with friends and family, probably at the beach.

What is one mental health strategy for managing hard days?

Depends on where you are in life, I think, when you're younger. My mental health strategy actually was to. Was to decide that I was gonna. I think what served me really well was pushing through really hard days, actually fighting through it. Our brains are just much better equipped to sort of grin and bear it as I get older. It's. It's. It's being more graceful with myself, spending much more time talking about my feelings with my wife and my friends.

Nice. And if you could. If you could have lunch with anyone, dead or alive, who would it be?

Oh, I am related directly to Benjamin Franklin. He. He is my. He is my fifth great grandfather. My. My mom is named after his only biological. His only. His only. Not biological. His only legitimate daughter, Sarah Bates Franklin. And I would love to spend time with Benjamin Franklin.

My daughter dressed up as. As him. I don't. That she did. I don't know if you haven't done this yet because you don't have kids, but the living statues. So she dressed up as him and.

Well, I was a kid once in California and did. And did in fact, dress up as Benjamin Franklin also, so.

That is so cool that that's your lineage. Like, amazing.

Yeah.

Anyway, last question. Wow. That's like a whole other podcast. Last question. If you had to describe your outlook in one quick motto, what would it be?

The world gets better all the time and optimism is not naive.

This has been so interesting. If my audience want to, like, book you for a speech or follow your work, what's the best way to do that?

Zach Cass Z A C k a s s.com and yeah, we have an advisory business and we have a book coming out May 31 called the Next Renaissance, the AI and the Expansion of Human Potential. And of course we get a lot of emails and I respond to most of them. So if people want to, they're welcome to respond to any of what they've heard today there.

Well, this has been so interesting and so fun. I hope you'll come back. Thanks for being here, everybody. Thanks for tuning into the Modern Customer podcast. Until next time, don't forget to subscribe to my YouTube channel or follow me on social media, including Instagram, LinkedIn X and more.