How do consumers really feel about AI? In this episode of CapTech Trends, CapTech CTO Vinnie Schoenfelder, discusses the findings of CapTech’s AI study, with CapTech Fellow Jason Snook and CapTech Manager Carrie Ryan. Tune in as we cover:
Read the full study here.
How do consumers really feel about AI? In this episode of CapTech Trends, CapTech CTO Vinnie Schoenfelder, discusses the findings of CapTech’s AI study, with CapTech Fellow Jason Snook and CapTech Manager Carrie Ryan. Tune in as we cover:
Read the full study here.
Vinnie
Hello and welcome back to CapTech Trends. Today we're going to talk about our AI consumer survey that we did this year. And with me, I have Carrie Ryan, a manager out of Denver on our CX team, and Jason Snook, who's been on at least one podcast before. We were trying to figure out how many a second ago. He's a research fellow who also calls CX home or has over the years and is in charge of a lot of the stuff we're doing with AI at CapTech and organizing kind of all of our independent research on that as well as our go-to-market strategies and such. So to kick us off, Jason, why don't you tell us a bit about the survey, what we did, why we did it.
Jason
And a little bit of background on that that I think is helpful is we track a lot of these tech trends for years. And what we're typically looking for, and we're tracking them today, what we're typically looking for is what I'd say kind of maturity and availability. These technologies mature to a point where they become relevant for our clients. We're tracking a lot of technologies that honestly our clients don't need to be thinking about right now. And so obviously, been tracking artificial intelligence for quite a while. We saw two things happen early last year. One was that that maturity level was rising to the point where we needed to start advising our clients to look at it. But the other thing that was happening, and you noticed this at the beginning of this year, everybody's writing a blog or writing an article about artificial intelligence. And if you read them, there's not much data. These are people's opinions on these things, and that's very valid. It's part of the conversation.
Vinnie
Either aspirational use cases or fear or legal issues.
Jason
All good. But the researchers, we like to be driven by data. And so we saw a gap there, or if we saw people using data, it was over a year old. And the way that this technology is advancing is really, honestly, anything older than six months, the consumer behavior has changed.
Vinnie
I noticed when ChatGPT started becoming a name, I heard it in restaurants, bartenders talking about it, or in a grocery store, people in line talking about people, non-technical fields. You heard it. A woman who's helping us decorate a couple rooms in our house was talking about it. The carpenter who was building some stuff for her was talking about it. So it went from something that a lot of our clients were using and that we were researching to really sort of a household name.
Jason
It's asked that way to mobile, where very quickly, there was a consumer recognition of this technology that we had to really understand well, because companies are going to do AI and we'll probably get into this AI for AI's sake. But what they're really going to be doing is designing products and services. And what company designs product or service for a customer, and they don't know what the customer wants, needs, feels about it? So it just felt like a real gap. So we moved pretty quickly to deploy something.
Vinnie
Great. And then Carrie, what types of questions were we asking? Was this limited to consumer research and if so, what aspects were we interested in?
Carrie
Yeah, we started off really the main sort of goal and the main purpose of our research was to understand consumer perceptions. So we really wanted to focus on those types of questions. Are people familiar with AI? Do they have experience with any sort of AI features? And then look at motivations around why people may be using it, what their overall sentiment was, if they found there were some tasks that they felt like AI could help in their day-to-day lives, things like that. So we really wanted to focus on the consumer side of things versus the kind of industry side of things.
We had one or two questions that asked about companies in particular, but it was again more from a consumer perspective, again, how consumers may feel that industries were doing from an AI perspective regarding communicating, the benefits, the risks, things like that. We had questions about whether consumers would recommend a brand if they knew that they were using AI. And we had one question where we did ask if people were in a decision-making capacity in their current company, what were some of the challenges that they had come across in trying to potentially incorporate AI technology in their company? So that was the one sort of company like question that we asked in the survey, but it was mainly consumer based.
Vinnie
I don't know if this is bias or just a gap... it's a question. Is there a bias or is there a gap in this way? If we ask consumers, many of which are going to be laypeople as it relates to understanding the intricacies of AI. We just did a podcast earlier today that's going to come out a week before a week after this one. One of the points I was trying to make is that good technology is frictionless. You don't see it, you don't know it's there, it's happening on your behalf and it's making the experience better and more immersive, but you're unaware of it. When you use your watch or your phone to open your hotel room door, you're not saying use Bluetooth to open door on your phone. It just does it.
So I guess my concern on the survey would be a lot of the valuable AI, a Tesla seeing three cars ahead, brake lights go on, and it breaks for you, your bank using fraud, waste and abuse detection and denying transactions on your behalf. My car has a microphone in the wheel well, so it knows when it's raining and it changes the driving characteristics. Most people don't know that's happening. So what they're responding to in the survey is higher friction AI, things they know that are happening to them and they're aware of them. Either one of you, do you think that changes the results in one way or the other from a survey perspective?
Jason
I'd affirm one part of that and then Carrie, you'll know the numbers a little bit more, maybe more accurately than I do. But any well-designed technology does tend to kind of fade into the background. If you design it well, it fits within the flow of a task that you're trying to do or a thought process or this or that. So one, we should hope we can do the same thing with AI that we've done with a lot of these other things.
Vinnie
We have in some cases and we haven't in others. I guess that's my question. Does that skew the questions? Because people will be answering the questions as it relates to the ones they're aware of, but not to the ones they're unaware of.
Jason
Yeah. And I thought that was one of the more interesting parts of the survey. And Carrie, you can fill in where I missed gaps. But we asked people first, are you familiar with AI? How do you feel about it? And then we asked them specific use cases and we asked them about voice assistance. We asked them about product recommendations. And what kind of surprised us, a little bit, coming out of it was that people were reasonably aware that Netflix was using AI to offer you different movies or things like that. They were reasonably aware that when they are online shopping, that even if it's dumb, even if you bought a baseball bat last year and now it's just showing you baseball bats, it was some intelligence behind it was trying to drive the behavior or the content I was being shared with. I don't know if there was a good example, Carrie, of just something that gave us the aha, like, oh, I think they do know AI is there.
Carrie
Yeah. We started out by asking again, are you familiar? And I think it was something like only 8% of people said they weren't familiar. But it would be interesting to do a follow-up study to find out are they familiar with it now because of ChatGPT and generative AI being everywhere. Were they aware a year ago when companies were still using AI? But to your point behind the scenes, they may not have known it exactly like you were saying with Netflix.
Vinnie
Well, you don't know what you don't know. Are you aware? I'm aware of what I'm aware of.
Carrie
So what was really interesting was we asked people, do you have experience? And it was actually split almost 50/50, yes and no. And then we asked, in your experience using AI features or applications, what did you find most valuable or impactful? And people said, voice assistance, personal recommendations, automated chat, automated customer service. So clearly, they have experience with these things and they may not have known that the kind of engine behind it was in fact AI. So they then recognize, okay, actually I have used this whether they knew it or not.
Jason
And then there's the point you made, Vinnie, that early on in technologies, they're not always designed very well, and so you take something like customer service automation. Well, they noticed it because it was probably bad. You've already talked to the robot and you really wanted to talk to the computer. You see these kind of feeble attempts at using automation. And so there's a little bit of that at play too. So where it's working well and where it's not. But I think overall that's where we were kind of surprised was that consumers seem to have a familiarity, not completely, but they were more familiar with the technology than we thought they would be.
Vinnie
And this goes back to friction. If I want to talk to a person, I've got to know the phone number, which is go to the website, it's buried because they don't want you to find it. You call, you're on a wait, then you have to go through, press four for this, two for this. Then you get the advertisement and then you're on hold. If I see a little chat button on, I don't care that it's a bot. If I can click it and type my question and get a response, so much faster than the other areas. So I think people are getting more comfortable with sort of the AI answers and not necessarily, it's now more cumbersome to talk to a person than a bot in many cases.
Carrie
Right. And using things like Alexa and Siri, you don't actually have to go into Google and type in anything. If you're in the car or something, you can ask Siri and you get a, for the most part, a correct answer.
Vinnie
So quick question for both of you. I'll start with you, Carrie. What surprised you on the results? What was unexpected?
Carrie
For me, the overall sentiment was much higher than I think we hypothesized. And when we looked at the data-
Vinnie
Meaning people having a positive reaction to it, not another fear-based reaction.
Carrie
Yeah. We asked people's overall sentiment toward AI and its impact on society. And we broke it down by very positive, I believe that AI has the potential to greatly benefit society, somewhat positive, AI can bring positive advancements, but there's some potential risks, then neutral, somewhat negative, et cetera. And when we looked at the data, it was, I believe close to 53% of the participants answered very positive and somewhat positive. And so that was very surprising. Obviously, we saw some concern around data privacy and things like that, which we should certainly talk about further. But that I think for me was the biggest surprise.
Jason
And I'll cheat. I'll tell you what didn't surprise me, but it was only after I saw the data.
Vinnie
That's not what I asked.
Jason
Haha, I know. It's dovetails to what Carrie was talking about though. It was interesting to see how open people were to the technology. So that was surprising in one sense. But it was interesting because what researched people for 20 years, people are very pragmatic. If there is a technology that can help me, that can make something easier or make it more efficient to do, we're generally very open to that. And the behavioral models have told us that for a long time. And so when it's at first like, oh, wow, they're very open to automating something or making something easier. But then I was like, well, in one sense that's always been the case that consumers are looking for something that adds value and they're adopting those technologies. So it was kind of a, oh. We were pleasantly surprised, but then I was like, that's how consumers behave.
Vinnie
Yeah. We talked about this briefly a while ago, and you used the word pragmatic because you're a nice person. I think almost Pavlovian, and I think I'm borrowing your word there as well, because talking about the two different aspects of that. And I say that because I see people who worry about the technology, who worry about, even outside of AI, they worry about privacy and data security and whatever else. But when they get what they want from something, it could be the stupid things on Facebook where take a picture of your face and see what it looks like when you're 30 years older and you can tell them that's going to models in China to be used against you.
And it's like, I don't care, it's funny. But you do it poorly, and now I'm a data privacy nut. So to me, people swing back and forth from I care about privacy and I care about how my data's being used when they don't perceive a high value. When they perceive a high value, all of a sudden their brand engagement and trust for the organization goes up. I don't think it's pragmatism because I don't think they're doing the math to say, "My data's still at risk and I want to balance that and work this out in a plus minus sort of way." I think once you start to get that positive response from it, it changes how you view the entire domain. Am I thinking about that right, or am I being too pessimistic?
Jason
No, you're giving a good nuance to it. And then Carrie, I'll throw it to you for the actual privacy numbers. But you said Pavlovian, or Maslovian? It is the calculus that's happening in us, we just don't know it. The research says that we're not consciously saying, "You know what? I'm willing to sacrifice my privacy for this picture of my friend's baby," or whatever.
Vinnie
Now she's got stars on her eyes.
Jason
Exactly. We're not actually saying-
Vinnie
Take my social security number.
Jason
Yes, "You may have all of my bank data." But it is actually what's happening at a visceral level if you can believe that. But I mean the privacy piece, Carrie did give them pause.
Carrie
It did, yeah. That is something where we asked a couple of sort of, do you agree with this statement type of questions? And those answers were, I'm not comfortable sharing personal sensitive data. That was, I think, the one that was the most heavily swayed in the sort of disagree category, if you will. So things like, to Jason's point, financial records, things like that, which is understandable. I think people are naturally hesitant to share that information or probably should be more so. But what I wanted to actually mention is something that we saw that was really interesting was even though people are hesitant to share that data, we did ask, have you ever experienced any negative or concerning issues related to AI? And overwhelmingly, the answer was no. We also asked, have you ever hesitated to use or adopt a product or service due to AI? And again, overwhelmingly the answer was no. So I think that's kind of interesting. They haven't had any negative experiences with this, but there is still that kind of preconceived notion that this could be potentially may be, quote, unquote dangerous for them to share that data.
Jason
Now, let's ground it in two use cases, and this was in the survey. We asked people how they would like to use AI. And so online shopping, that was top, that probably makes sense. The next one was healthcare monitoring. And so I'll juxtapose that with another in a second, but it was this idea, and maybe not a conscious one, but how much of your health data I'm going to need to be able to use AI to do that. And I think for consumers, we're reading into the data that in some cases, yes, if this is going to help improve my health outcomes, that's one thing. An interesting juxtaposition that I'll be interested in, you can poke holes in this, is somebody asked me the other day whether or not AI could do their taxes. And I said, "Yes, if you dump all of your financial data into GPT," and not going there. So I'm trying to understand the nuances between these different use cases and the value judgments consumers are trying to make because-
Vinnie
Well, I think when things are anonymous, it helps. Also, I would debate the health one, if you want to make significant predictions about my long-term health, yeah, you need a lot of data. However, you can do a lot of interesting things with a very small amount of data.
Jason
That's true.
Vinnie
From me, but there's a huge pool of it. For instance, when elderly fall and break a hip or sometimes your hip breaks and you fall sometimes, I don't know if it's a chicken or egg kind of thing. Very serious, as you get older. If your phone in your pocket or a smartwatch, if you just measure walking cadence, step length, how long it takes to get up and down stairs, are you favoring a stride side, et cetera. Small bits, very, very small bits of just walking data and you put that into a model, you can predict that a person will have a fall within a week because changing how they're walking. So that information isn't blood type, it's not DNA, it's a very small amount of behavioral data, but at scale can have a big impact.
So to answer your question directly, I think to the extent that information can be tied back to you and is personal or financial, people cringe. But if it's very basic information and anonymized and the value is perceived as high, then that goes into that Pavlovian or pragmatism, whatever that model is in our brain. One of the things, since nobody asked, what I found interesting in the survey, one of the things that jumped out at me, I'll just offer it up, was to your point about people having a positive impression of it. They actually expect it now from the brands that they're working with, the companies are working with, the apps that they're working with and they're expecting to see it. So if it's not there, that could be an issue for companies who could lose customers to other sites that are more highly personalized or have better workflows because it isn't there. So it's not that they don't mind it, they expect it. How are you guys seeing companies communicate? Are they communicating that it's there or are they just making things better and then you experience it?
Jason
Well, we should get into how they're communicating privacy, because that's important. But what that made me think of is you were talking about friction earlier and the idea of using AI to remove friction, to add insights, to automate something. And it's answering your question because I think that, and we did have a question, Carrie, where we said, if a company was using ai, would you be more likely to use it? And the answer was skewed yes. But I think there's also going to be these subtle improvements to products that remove friction, automate something that wasn't before, and it's going to add value.
The piece I'd add to that, and then I'll share the mic, is around, we really are telling our clients there's a first mover advantage to this one. If you took a couple years to migrate to the cloud, you probably should, but you're probably not going to face a huge competitive advantage. But the models that companies are building that are removing the friction, that are adding insights to their products, they're getting better. And if you wait six months, that means theoretically your competitor's model is six months better than yours.
Vinnie
I would say better than that because I think as we face more and more legal issues, the quantity and availability of data is going to get smaller. We're seeing this now with Reddit. They had open APIs and people were scraping, scraping, scraping and scraping, and they put paywalls up for large numbers of API requests. So not that I'm suggesting that people build these models to go out and scrape the internet as much as they can. But to the extent that there are available data sets for you to learn from, those might become smaller or more expensive as people get wise as to how data's being used and the real value of the data that they have.
Jason
Absolutely, and it was funny, Carrie, with that question. If you take it from a Pavlovian response to like if I ask you point-blank, is privacy important to you? That question, I liked how we asked about privacy. But on its head, there's a way that you're also supposed to answer that question. So there's a little bit of skew there. But still, it was a lot more neutral than I thought it would be. It definitely gave people pause.
Carrie
Yeah, I totally agree there. I think the other thing, to your question, Vinnie, about how well do consumers feel that companies are actually communicating, we asked this in a way to be almost all-inclusive. So our companies communicated benefits and risks and potentially limitations of AI, and that was where people pretty overwhelmingly said that companies were not doing a very good job with this, which I think is interesting when you think about people's then answers to what they're comfortable sharing, whether that's sensitive data or their shopping history, things like that.
So I think there's a big opportunity for companies to increase their communication, not just around what they're doing in terms of securing people's data and communicating the risks. But what they're also doing to what the benefit is for the consumer, that's also very important. But certainly there's a big opportunity for companies to do a better job around exactly what they're doing, not necessarily that people are interested in knowing the technology behind how they're really doing this, but the advances that they're doing to secure people's data. People really want to know that, and I think that's going to help build trust and build potentially brand loyalty if that's a concern for people. I think that's an area where companies can do a better job.
Jason
And to emphasize one distinction you were making there, Carrie, that I thought was really interesting is we asked two questions. We asked one, would you trust them with your information, call it more innocuous, your shopping behavior, that kind of thing. And then right on the heels of it, we asked, would you trust it with your sensitive information? And that's really where you see it skew off and all of a sudden, adding that word was really important. And it speaks to some of the nuance you were talking about, Vinnie, where if I could trace it back to me, if it's something I really don't... do I care what people see if people knew what I shopped for on Amazon? Me personally, no.
Vinnie
But do I want one of those car insurance devices plugged into my car so they can see how I drive my car? No. I do not want that.
Jason
Do you want to see how many orange lights I've ran? So not there.
Vinnie
If that was completely anonymous and they were looking at it at a statewide level, and I trusted that it was, and that would help save lives, now it's a different conversation. I remember when COVID came out, having those apps where you could track say, "I have COVID," and if you were next to somebody else who had COVID and all that. And I looked at the technology on that, and that was truly anonymous, and I was comfortable with that and I downloaded it.
So yeah, it really comes down to understanding that line, seeing that line. Kind of wrapping up or putting a bow on this from a consumer standpoint, if you've got customers out there, and by the way, this is available for anyone to come check out, there's no paywalls. Just come out to the marketing site at CapTech and you'll see it. Customers are comfortable with it, they want it, they expect it, and they want it around things that improve efficiency of transactions, personalization that we know we can remove steps in the workflow to get things faster, things that are more immersive, and actually safety. When it comes to fraud, waste and abuse and other types of things to make sure that your transactions are more secure.
Did I wrap that up well or would you add any other sort of consumer expectations on there that I missed?
Carrie
Yeah, I think that's a great summary. I think people are overwhelmingly, the use cases that we saw throughout the survey were around how can I make my life more efficient? They're comfortable with the idea of AI helping them in their daily life. So those are really the use cases that we saw throughout the survey in terms of the want and the need from AI. I think the other thing that I did want to just quickly touch on, which we sort of touched on a little bit earlier, and I know Vinnie, you mentioned this in terms of exposure, really can accelerate adoption. So if we can just get those people that are maybe a little bit hesitant to even try, whether it's ChatGPT or another maybe generative AI feature that's more front-end consumer-facing, the adoption skyrockets. So I think that was another really interesting piece. I know for me, when I first tried ChatGPT, I think my own preconceived notions where I was a little bit hesitant and I tried it and now I'm so reliant on it.
Vinnie
It's an interesting thing from an app or website perspective, how can companies, I don't want to say gamify, but almost sort of encourage experimentation so that they can see... I totally agree with you. The woman who was helping us redesign some rooms in our home was, "I'm scared. This is scary. I don't know what to make of this." And I sat down in front of ChatGPT, and she's a designer, and I said, "Well, what would you like to design?" And she's like, "A 2,500 square foot beach house with," blah, blah. And I was typing as she was saying it, and I hit enter and it was great. It gave her a full laid out floor plan, and she was like, "Oh my gosh, that's good."
But then as she looked at it more closely, she was like, "This is really amazing, yet there's no front door." And so it's like, okay, now we've got to go back and correct some things. But from an inspiration standpoint, all of a sudden now she's asking more and more questions and getting it. You could see the excitement, the fear go away and the excitement step in. So I think getting people to experiment with that kind of stuff is important. Jason, any closing thoughts from you?
Jason
Yeah. I think what we'll see over the next year, we'll be looking for those kind of one-off applications of AI. You'll see it's sprinkled in experiences and it may take a year for companies to get ready is when they run AI full on through the entire experience where it's having second and third order effects on the customer experience-
Vinnie
What do you mean by second or third order effects?
Jason
So kind of this idea. So you take your banks, being able to use your data to come up with an entire financial plan for our hospitality companies to be able to come up with an entire travel itinerary and maybe even book it for you is really, I think, where these things, we're a ways off from that. But I also think that we're closer than people think. And it's the companies that first roll out these second, third order kind of AI experiences that I think are going to catch headlines and really get in pole position. And that's why I think there's a first mover advantage.
Vinnie
Last question. Are we going to run this survey again, and if so, how frequently?
Carrie
I know the want is to run a more sort of industry-based survey, and that's in the works I would say, if you agree with me there, Jason. And then we were also talking about potentially running this exact same survey in six months. Just see if sentiment or perceptions or anything like that has changed, and so we can have that sort of trend data to look at.
Jason
It's moving so fast, we're going to have to. And I'm excited to do it. So there'll be much more I think by the end of the year.
Vinnie
Well, Carrie, Jason, thank you so much for joining on me today. A great conversation. For those listening, the survey's out there. Is there a way on that page, Jason, for them to get in touch with you, or Carrie or someone else if they got questions?
Jason
Yeah. Actually, I think CapTechconsulting.com/AI will take you there.
Vinnie
Will take you there. Okay.
Jason
I think our head bubbles are around the bottom of the page, so either that or LinkedIn.
Vinnie
Great. Yep. So if you guys have more questions, just reach out and we'll follow up. Thanks so much for listening.
The entire contents and design of this podcast are the property of CapTech are used by CapTech with permission and are protected under US and international copyright and trademark laws. Users of this podcast may save and use information contained in it only for personal or other non-commercial educational purposes. No other use of this podcast may be made without CapTech's prior written permission. CapTech makes no warranty, guarantee or representation as to the accuracy or sufficiency of the information featured in this podcast. The information, opinions and recommendations presented in it are for general information only, and any reliance on the information provided in it is done at your own risk. CapTech makes no warranty that this podcast or the server that makes it available is free of viruses, worms, or other elements or codes that manifest contaminating or destructive properties. CapTech expressly disclaims any and all liability or responsibility for any direct, indirect, incidental, or any other damages arising out of any use of or reference to reliance on or inability to use this podcast or the information presented in it.