Awin is the all-in-one platform for smarter partner marketing. With award-winning technology, expert support and a global network of 1 million partners, Awin helps brands discover, manage and scale high-performing partnerships that drive measurable growth. Learn more.

Marcus Johnson: [00:00:00] Today's episode is sponsored by AWIN, the world's leading brands grow through partnership marketing with AWIN's enterprise technology, AI-powered insights, and expert US regional support. With 30,000 advertisers, one million partners worldwide, and more than $20 billion in partner-driven revenue generated last year, AWIN helps brands turn partnerships into growth.

Marcus Johnson: Learn more at awin.com/us. Join EMARKETER on September 15th in New York for the Future of Digital Summit. It's our biggest event of the year. You can hear exclusive research and executive insights on AI visibility, commerce media streaming, and agentic commerce, and leave with a roadmap for what comes next.

Marcus Johnson: Click the registration link in the show notes for tickets.

Marcus Johnson: Hey, gang. It's Monday, August 24th. Stephanie, Nate, and listeners, welcome to Behind the Numbers, an EMARKETER podcast made possible by AWIN. This is the show that helps you keep up to date with everything media, marketing, tech, some other things in about 20 minutes or so. I'm Marcus, and joining me for today's conversation, we have two [00:01:00] kind of New York-based people.

Marcus Johnson: Now our VP of insight, Stephanie Paterik joins the show. Welcome.

Stephanie Paterik: It's great to be here. Long time listener, first time caller.

Marcus Johnson: Pleased to have you. Joins us from our New York studio, and we also have our AI expert, Principal Analyst, Nate Elliott.

Nate Elliott: Greetings, Marcus. Long time caller, first time listener.

Marcus Johnson: We know.

Marcus Johnson: Uh , today's fact.

Marcus Johnson: Okay, so the word quarantine, oof, it's a tough one. Uh, i- essentially it's linked to 40 days. So the word originated in the 14th century during the Black Death plague to protect coastal cities from infection. The Republic of Ragusa, probably said that wrong, now, uh, the city Dubrovnik in Croatia, uh, and later Venice, required arriving ships to sit at anchor isolated offshore before passengers or cargo could come ashore.

Marcus Johnson: It started as a 30-day [00:02:00] isolation period called a trentino. The Italian trenta means 30. Until authorities, uh, extended the isolation period to 40 days to ensure the disease had fully run its course. The Venetian language changed the term to quarantina, uh, from quaranta, meaning 40, and this evolved into the English word quarantine.

Nate Elliott: What if it had been 70 days?

Marcus Johnson: They probably should have- They probably should have made it 70. I don't know what it would have been called. They probably should have made it, uh, I guess sep- sep- sub- sepati- something with sep in it. Uh, they should have made it 70 days because, um, the... and this wasn't instituted fast enough because in just seven years the Black Death killed close to 200 million people, which at the time was about half of the world's population.

Marcus Johnson: Um, it's kind of like Avengers. You know when he clicks his fingers and, like, half the world... That was literally what happened with the Black Death. So 70 days, not a bad idea, Nate, to be honest. [00:03:00] It's a shame you weren't around back then. They would've been like, "40 is not enough."

Stephanie Paterik: It's not working. Marcus, this is a bummer stat.

Stephanie Paterik: We can only go

Marcus Johnson: up from

Stephanie Paterik: here.

Marcus Johnson: I know. I'm so sorry. Well, I did also read that in five billion years the sun will implode.

Stephanie Paterik: Oh.

Marcus Johnson: So even if we do make it that long, we're, we're gone. It's been pretty... yeah, it's been pretty dark.

Nate Elliott: This is- Well, let's talk data. On a lighter note-

Stephanie Paterik: Yeah.

Nate Elliott: On, on a lighter note- You could have had that one

Nate Elliott: someone in the control room needs to tell us what percentage of podcasts Marcus mentions the Avengers in, because I- Ooh ... I feel like it's a non-negligible number.

Marcus Johnson: Most is the answer to that. Yeah. All right. What a rough start. Sorry about that, gang. Anyway, today's real topic

Marcus Johnson: How people use AI. All right, gang, so today we are asking how are people using AI. But, um, Stephanie, let's start here because when we were discussing what to discuss, uh, Nate, you made a really interesting point that, um, uh, uh, adoption, uh, could [00:04:00] be thought about, uh, differently. Uh, so, uh, uh, Stephanie, why is it important for folks in our world in particular to perhaps think about AI adoption differently than perhaps we have been?

Stephanie Paterik: I love this question. I love that we're starting with the why, because I know we've got a lot of great data and insights to share, but at the end of the day, why does AI matter to marketers at this moment in time? When I think about how many people are currently thinking about AI, I think that they are caught up in the hype.

Stephanie Paterik: I think it's easy for all of us to get caught up in the hype. You know, you go to any industry event, and every panel is about AI. Um, everyone has like worked AI expertise into their LinkedIn bios and profiles. And, you know, obviously we have a lot of AI businesses starting up. And that creates this feeling that there is this wholesale transformation happening all at once.

Stephanie Paterik: And I think the reality, and I think a more useful way to look at it, is that it is, it's a more subtle [00:05:00] change that is going to play out over a longer amount of time. And I think it may be a deeper change than we expect, but more subtle and more long-lasting. Um-

Nate Elliott: Yeah ... I,

Stephanie Paterik: I mean, I, I'll just say too, I, to me- Please

Stephanie Paterik: one of the most fascinating things that I'm seeing about AI right now is that I think it's really emerging as a partner for us in making decisions. And Nate has some really great research about use cases and how people are, are using AI. But when I step back, what I really see is people who need help making a choice.

Stephanie Paterik: And it might be at work. "I have a presentation due. I need help deciding how I'm going to organize this presentation." It might be, "I, my spouse just did something to piss me off, and I need to decide how I'm going to respond." Um, it might be, "I have a gift card to spend at a store I haven't been to before, and I need some help narrowing down the options."

Stephanie Paterik: So the emergence of this new [00:06:00] step in, in, in the decision-making process and this new partner in dec- decision-making is the piece that I think there's a huge opportunity for marketers to key into right now.

Nate Elliott: Mm-hmm.

Stephanie Paterik: But Nate- We- ... I'm curious. How about, what do you think?

Nate Elliott: Well, first we wanna hear about that second example once we're off air.

Nate Elliott: Um, it's, no, it's, it, it's a great point. And, and the thing I'd add is that- AI adoption is happening unevenly. It's happening unevenly in terms of different groups of people adopting AI with different speed, um, using it at different frequencies, using it for different reasons and, and at different depth, right?

Nate Elliott: So some people use AI for lots of different reasons. Some people use it just for one reason. Some people use AI 12 times a day, and some pe- some people use it 12 times a year. Um, all, all of that looks different when you look at different groups of people, and it's really important to know that, especially if you're a marketer hoping to leverage this as a marketing channel.

Nate Elliott: It's important to know if you're [00:07:00] any organization hoping your workforce, your employees will adopt these tools and use them to get better at their jobs. And the other thing I'd add is AI, like every other technological innovation I've seen in the last 30 years, whether it's mobile, whether it's social, whether it's the metaverse, whether it's the internet in general, um, suffers from shiny object syndrome.

Nate Elliott: Everyone's looking at the tools. Everyone's looking at the new model releases that are happening literally a couple times a week, uh, if you look across all the different players, all the new features and functionality, uh, all the new benchmarks. And it's so easy to focus on the shiny object instead of what people are doing with these tools, and how, as marketers in particular, we can leverage these tools to reach people.

Nate Elliott: And I think that's why it's important not just to look at who's winning the race between ChatGPT and Gemini or, you know, what benchmarks are being hit by which versions of which models, but [00:08:00] actually to understand the human motivations that drive people to use these tools in the first place. Yeah.

Marcus Johnson: You have this quote in, in your research, um, uh, this line in your research, uh, that personal motivation is essential to any behavioral change, um, according to Stanford University's Fogg Behavior Model.

Marcus Johnson: Um, and they, in the piece, you're basically saying that, like, websites have come and gone, but people still cite the same top reasons for going online today as they used to, which is finding things and connecting with people, um, as G- GWI cited there. Um, I wanna touch on something you said in terms of, um, the shiny new object and, uh, h- how people are using, uh, AI.

Marcus Johnson: But I guess also we're talking about the people who are using AI. We, we have to remind ourselves that not everyone is using AI, and not everyone is aware of AI. And the ones that are aware are aware at different, uh, to varying degrees. Pew Research has some data. They said in late 2023, so that's three years ago, nearly [00:09:00] 30% of Americans had never heard of chatbots.

Marcus Johnson: That number fell to 10%, um, uh, in a couple of years. Put the other way, that means 90% of US adults report having heard of it. However, when you break that 90% down, 44% say a lot, 53% say a little. So it leaves a lot of folks with a pretty unfamiliar, um, uh, understanding of the technology. Of the people using it, Na- well, how many people are using it?

Marcus Johnson: Let's start there. How, how many folks do we calculate are using AI right now?

Nate Elliott: 152.9 million.

Marcus Johnson: That's quite precise.

Nate Elliott: Yeah, it is. Okay. That's, so that's our, um, our, uh, EMARKETER forecast for how many people in the United States will prompt a generative AI tool like ChatGPT or Gemini at least once a month this year.

Nate Elliott: 152.9 million, that's just under half of the US online population. So- Okay ... you know, Stephanie made this point before, it feels like this wave [00:10:00] crashing over us, and everyone's using these tools, and in fact, even by the end of this year, the majority of people who use the internet in the US will not be active AI users.

Nate Elliott: Only 48.8% of them will use AI on a monthly basis or more often. Mm-hmm.

Stephanie Paterik: And Nate, I think that's a great distinction too, because there's the bucket of people who are familiar with it, who've tried it, and then there are those who are active users, and those two groups look really different.

Nate Elliott: They do, and, uh, you know, the audience who's listening to us right now are am- amongst the heaviest users of AI.

Nate Elliott: It's a really interesting conversation to have with people in this industry. If, if the entire US population looked like EMARKETER's clients or audience, then 99% of people would be using AI, and Claude would probably be the most popular tool. And in fact, neither of those things is true.

Marcus Johnson: Um, Nate, when you said y- y- use AI, uh, [00:11:00] y- we're talking about, uh- Opening up, uh, um, a, um, uh, an AI model, a ChatGPT, uh, and, uh, and, um, a Gemini, uh, as opposed to searching for something and an AI overview popping up, correct?

Nate Elliott: Yeah. So, um, so we're distinguishing in our forecast, uh, we talk about AI users, the people who actively prompt an AI model. So, uh, I mean, AI overviews are by far the number one way people engage with AI content, right? M- more than two and a half billion people see an AI overview on Google alone every month.

Nate Elliott: That's far more than the number of people who use ChatGPT or Gemini or AI mode, which is Google's AI chatbot built into the search results page, different than AI overviews. By far the largest way people interact with AI is through Google's AI overviews, and they're not asking for those AI overviews.

Nate Elliott: They're going to Google or they're going to their browser's search bar or, [00:12:00] uh, or URL bar and typing in a search query, getting a, some form of a traditional search results page, and Google themselves are deciding to put an AI summary of the search results at the top of that page. So that, that's how the largest number of people are engaging with AI, but we're not counting that.

Nate Elliott: When we talk about the fact that 153 or so million people in the US are gonna use AI this year, that that's about a little under half of the US online population, we're not talking about just seeing an AI summary that a company decided to put in front of you. And by the way, that goes for other AI summaries.

Nate Elliott: It goes for the AI summary of reviews that you see when you look at a product page on Amazon or any other example you can think of, of AI summaries. What we're talking about is people making an affirmative choice to go into an LLM or an AI chatbot and prompt it to ask it a question to look for an AI response from a tool designed to give you that AI response.

Nate Elliott: [00:13:00] So that's using ChatGPT or Gemini or Claude or Perplexity, Groq, uh, any of those tools.

Marcus Johnson: Uh, you mentioned a number there. There are, uh, obviously more. Which platforms are most popular at the moment, and how has that changed over, over the couple, the last couple of years?

Nate Elliott: Um, of course, there are a lot of ways of defining most popular.

Nate Elliott: Uh, the- Yeah So, um, the way that, that I, uh, most commonly talk about it is how many people are using these tools on a regular basis. And, and listen, that doesn't mean how much usage do they get. We can look at it that way as well, right? So you might have the same number of people- Mm-hmm. Interesting ... using Google's Gemini model as you have using ChatGPT, and, and actually that number's very close, right?

Nate Elliott: Um, um, Gemini, uh, Google just recently, uh, announced that Gemini had reached one billion monthly users. Not long before that, they revealed that AI Mode had reached one billion monthly users. Uh, we're still waiting for ChatGPT [00:14:00] to confirm that, uh, that they have a, a billion weekly users. There's been some confusion over an, a cryptic reference to a billion users in a press release or a blog post not long ago, and, and it wasn't labeled as weekly or monthly users.

Nate Elliott: So, um, they're confusing us all right now. Um, but the, the numbers are pretty close. But, but even if Gemini had the same number of monthly users as ChatGPT, it doesn't mean they have the same number of weekly users or the same number of daily users, or that the time people spend every day- Yeah ... or the number of prompts or turns that they do are the same.

Nate Elliott: So right now, ChatGPT has a slender advantage over Gemini and AI Mode, uh, both individually and also if you combine those two. ChatGPT still has a slender advantage over those, uh, models from Google in terms of monthly users in the US. But our forecast says that Google's two models combined, uh, Gemini and, and AI Mode, are going to [00:15:00] overtake ChatGPT in terms of monthly users in the US in the first quarter of 2027.

Nate Elliott: Hmm. Despite that, it'll take some time for them to overtake ChatGPT in terms of usage, in terms of volume of prompts or conversations. And, uh, we have this great data that you can see on screen that shows that, um, ChatGPT still has more than half of worldwide gen AI web traffic. Uh, so even though its lead is going away very quickly, and even though this number, the, the percentage of global AI web traffic that they claim, has gone down immensely over the past couple of years, they do still own more than half of it at last check in terms of just the number of conversations, the amount of traffic and pages and responses and prompts that are going back and forth.

Marcus Johnson: Yeah. Yeah, it's, I said popular, but it, Stephanie, it is interesting how folks are, are going to, and which engagement metrics people pay more attention to 'cause [00:16:00] to, to Nate's point, um, is it monthly, weekly, daily? Time spent you could argue is, given the conversational nature, um, of these models is perhaps even more important than it ever has been.

Marcus Johnson: Um, what part of this engagement or with AI conversation interests you the most?

Stephanie Paterik: What's interesting is that we see right now people are spending on average 16 minutes a day with, with AI, within LLMs. Mm. And so that's, that's, I mean, it's definitely popped up into our time use chart, right? But it's not a huge amount of time.

Stephanie Paterik: When you think about how much time you spend on social, how much time you spend streaming television, 16 minutes is a, is a pretty, you know, compact amount of time. Um, it is possible though, you know, we have some channels that are, are low time spent but high intent, so someone is coming with- Mm ... a particular question in mind or a problem that they want to solve, and they might not need a lot of [00:17:00] time, uh, uh, for it to be useful for them and for it to influence them, which I think is super interesting.

Nate Elliott: Yeah. And that's 16 minutes. We, Steph, we sat down and looked at this recently, right? It's 16 minutes per day out of what? 15 hours or something like that of total time spent- Mm ... with media in a day. Yeah.

Stephanie Paterik: I think 14. Yeah. Mm.

Nate Elliott: So it's, I mean, 16 minutes out of 14 hours is vanishingly small, but, but as you say, if the intent is high, that makes it valuable not just to the consumer, but to advertisers as well.

Nate Elliott: Mm-hmm. And that's what we're starting to see folks lean into in terms of GEO and ads on ChatGPT and so on.

Stephanie Paterik: Yeah.

Marcus Johnson: Yeah. I imagine, I mean, it is a small amount of time relatively, but given that people have been using these models for a couple of years, I imagine that the, the growth in, in minutes is, like in, in a couple of years, then the amount of time people spend with AI is gonna be, um, significantly greater.

Marcus Johnson: Is that fair to, to assume?

Nate Elliott: [00:18:00] I will say this, it, it's gonna get more confusing, um, rather than less confusing because- Interesting ... the whole game for Google is to show people that they don't need to go somewhere else to get AI responses, right? When you look at AI overviews, when you look at AI mode, what they're doing is trying to show people this information-seeking behavior that you have had just imprinted into your brain over the past 30 years.

Nate Elliott: I go to my browser bar and type something in, and I get Google results, or I go to google.com and type something in, and I get results. That's the way the vast majority of the internet finds information the vast majority of the time, and what they want is to make sure that as AI grows, it doesn't disrupt that behavioral pattern because if it does disrupt that pattern, well, Google made- Something like a quarter of a trillion dollars in search advertising in the U.S.

Nate Elliott: alone last year. So it's a lot [00:19:00] of money that could go away if people start changing their behaviors when they want information. And what Google wants to do therefore is to make sure that people can still get the information they're looking for in this new format that they might prefer without changing those behaviors.

Nate Elliott: That's why there's an AI overview summary at the top of so many traditional Google results pages. It's why there's an AI mode tab available at the tops of those pages as well. And what that means is, as we look to count the minutes people spend with AI versus other types of media, well, AI and search are merging.

Nate Elliott: They're becoming the same thing. Uh, it will be exceedingly rare for someone to find traditional search results that don't have an AI overview or potentially that don't deposit you directly into AI mode or a full-fledged chatbot conversation. In a couple of years, search and AI are going to look very similar and Us trying to count it will [00:20:00] just get a lot harder.

Marcus Johnson: Good luck.

Stephanie Paterik: Um, another thing too just that struck me as you were talking, Nate, on the competitive landscape, you know, we look at 16 minutes time spent on a channel, but what's very interesting, too, is how people are spending time on different LLMs. I think that they are emerging as strong in different areas, right?

Stephanie Paterik: So you have people turning to Claude. I'm gonna make some overgeneralizations, but Claude is becoming known for a place that you can turn to for work, right? Uh, uh, Gemini, I think because of its relationship with search, you know, is, uh, is, is... I perhaps see it as like a mass market. You know, it's easily accessible, um, easier, e- e- lower barrier to entry, right?

Stephanie Paterik: Um, ChatGPT, perhaps that's who I'm talking to about, you know, the, the, the, the conversation with my husband about who needs to unload the dishes. And by the way, if you're watching, Chris, I love you. Um,

Nate Elliott: But, but Chris, unload the [00:21:00] dishes.

Stephanie Paterik: Yeah. But, uh, Nate, I'm, I'm curious what you think about, you know, that horse race, and is there room for all of them?

Stephanie Paterik: Is this going to develop like the world of streaming where I've got my Netflix subscription for my K-dramas, and I've got, you know, my Disney subscription for my kids and, um... Or, uh, uh, are we moving to a, you know, sort of one LLM to, to rule them all scenario? Great

Marcus Johnson: question.

Nate Elliott: Yeah. I, I think it, I think it depends on whether you're talking about getting information or getting things done.

Nate Elliott: Th- there's certainly room for multiple different technologies in terms of getting things done. I don't think there's a lot of room for a lot of different technologies, but y- you know, you talked about Claude being a place where people get work done. Copilot, of course, is a place we see in the survey data a lot of people are doing work.

Nate Elliott: And it may turn out that some combination of, you know, Claude or Claude Code and Copilot end up dominating how people use [00:22:00] AI in general work and office settings. And then on the asking questions side, uh, I, I can only imagine that this will end up being a one or two horse race the way that search has been a one or two horse race.

Nate Elliott: People forget that- The reason Google won initially was it was just a lot better than everything else. When Google launched almost 30 years ago, 27 years ago or something like that, there were a dozen different search engines and, and the, the crown for which one was most popular changed every six or 12 months for the first few years.

Nate Elliott: Hmm. And Google came along with its PageRank algorithm that was meaningfully better, that gave meaningfully better responses and links to what people were searching for, and that's why it took the crown. The reason it kept the crown was because it introduced a decision layer that everyone else was slower to adopt.

Nate Elliott: And the whole thing was, you know, Google launched Maps search and Shopping search and News [00:23:00] search and Video search, and then watched very few people go and use these tools because people just wanted to go to one place to ask questions, and the decision layer that Google built in allowed people to just use a single search box for everything.

Nate Elliott: And Google figures out, is this actually a Maps question? Is it actually a Shopping search or a News search or a Video search? And people don't actually see 10 blue links that often. Even before AI Overviews, people would see sometimes 10 blue links, but a lot of times they'd see a Maps result or a Shopping result or a News result or a Video result when they just use the general Google search engine.

Nate Elliott: Yeah. It was that decision layer that said to people, "No matter what you need, you can come to this one place, and we will figure it out and get you headed in the direction that you need to go in." You know, we think that, that Google's AI models are going to be the ones that dominate the consumer side of AI, and that leaves the enterprise side [00:24:00] or the workplace side.

Nate Elliott: And there, you know, Claude and Microsoft and, and OpenAI seem to be in a pretty decent horse race at this point. But I don't think we're gonna see people using... choosing to use six different models for six different reasons at any point. Hmm.

Marcus Johnson: Hmm. We talked a bit about, at the very beginning about the kind of why for adoption, discussed how many people were using it, the where, being which platforms.

Marcus Johnson: Um, let's, let's move to the what. Uh, Nate, to, to tackle this in your research, you put together, um, what you call the building blocks of AI, uh, framework. Uh, tell us a bit about, um, uh, how, how this came about a- and, and what this looks like a- and, um, it's, um, and what's going on in terms of what people are using AI for.

Nate Elliott: Yeah. It, so it's our attempt to address that shiny object syndrome I mentioned earlier. If you look at the top 10 websites this year compared to the top 10 websites a decade ago, half of them are different than they were, and [00:25:00] yet the majority of them are focused on finding information and connecting with people.

Nate Elliott: So i- motivations don't change very quickly, even as technologies do change really quickly, and we're using that as an instructive concept as we look at the evolution of AI. Uh, you know, what we're seeing right now is that the technologies are changing remarkably quickly. Uh, again, uh, you know, ChatGPT introduces a couple of new models or significant updates to their models every month.

Nate Elliott: Uh, you know, and we see that from Anthropic, we see that from Google. The technology's changing very, very quickly. But what we don't expect is gonna change nearly as quickly is the human motivations for people to l- to use these tools, to turn to these platforms. And so we ran a survey earlier this year.

Nate Elliott: We're gonna do it every six months, so we're actually sitting down to meet soon on the next wave of this survey. Um- Like right

Stephanie Paterik: after this, I think.

Nate Elliott: Yeah, very, very soon. And so what we're looking at [00:26:00] is, you know, why are people using these tools? Are they using them out of curiosity to find information or facts or explanations?

Nate Elliott: Are they using them to get things done and make their personal lives work better, or to get things done in their work lives to, to make their jobs smoother and make them better at, at what they do for a living? Are they using it for fun, entertainment, for playing? Are they using it to maybe find and purchase the things that they need in their lives for shopping-like purposes?

Nate Elliott: Are they using it for some form of connection, actually reaching out to these tools for conversation or for connection, or even for things like therapy? Um, and we built this model called the building blocks of AI that, that looks at these categories of asking and doing and playing and working and shopping and connecting, and shows what percentage of the online population is turning to AI tools each week to do each of these things.

Nate Elliott: And, and the really interesting thing is not just how these stack up for the general population, although I find that somewhat fascinating, it's to [00:27:00] look at how those numbers are sometimes radically different by generation, by gender- Yes ... by income, by all these different cuts of, of user demographics and psychographics and behaviors.

Stephanie Paterik: Nate, I love that you brought up generational differences, gender differences, because I think that there's really interesting data emerging, uh, exactly in that spot. And when I look at the shopping use case, the shopping block, for example, um, there's some data that 61% of Gen Zers have used AI to help with a purchase, and only 15% of boomers.

Stephanie Paterik: And so if you are a marketer, if you are a brand, uh, uh, looking, you know, to create attention, to create conversion, uh, to ultimately impact your bottom line, a lot has to do with who you are trying to reach. And, uh, I think, you know, there's, uh, I also read something recently about it's gonna be interesting to see what happens with Gen [00:28:00] Alpha once they learn how to read and write, how they use AI once-- They just gotta learn to read and write, and then they're gonna come in and, you know, maybe, uh, using it, uh, in new and interesting ways, too.

Stephanie Paterik: So, uh, yeah, su-super interesting, and it's why we can all be having a different experience with AI and a different impression about how revolutionary it is. Uh, yeah. Marcus, how are you using it?

Marcus Johnson: AI?

Stephanie Paterik: Yeah.

Nate Elliott: Mostly to find facts of the day.

Stephanie Paterik: To, to, to read up about, uh, quarantine.

Marcus Johnson: It's become most of my life. Um, but I, I think personally, um, it's a good French teacher.

Stephanie Paterik: Ooh.

Marcus Johnson: If you need someone to go back and forth with to learn how to, um, speak a language, uh, some part of the hard, uh, the, some of the hardest, um, parts of learning a language for me have been getting the confidence to, to say something and then when someone says, says, says something back, uh, taking a second to, to process what they've said before they walk off or [00:29:00] decide that you should speak in English because they're better at English.

Marcus Johnson: So, uh, yeah, I found it very, very helpful, uh, there. Um, but it is fascinating to look at the different, um, generational, uh, differences. Um, I, I've like a billion questions, but I can't get into this. Yeah. We've gotta, I've gotta wrap because-

Nate Elliott: Marcus, by the way, 9.8% of US online users say that they've used AI to translate something from another language in the last month.

Marcus Johnson: Uh, j- you said 9.8%?

Nate Elliott: 9.8%, yeah.

Marcus Johnson: Oh, wow. Um, I'm surprised it's not... Yeah, I'm a bit surprised it's not higher. Um-

Nate Elliott: Well, remember, only 50% used it last month, so-

Marcus Johnson: True ...

Nate Elliott: that's, you know- True ... uh, you know, one-fifth of people who actually use AI at all.

Marcus Johnson: Yeah, that's a good point. Uh, listen with this, Nate, um, what's one thing marketers should be thinking about when it comes to their customers' AI adoption?

Nate Elliott: They should be thinking about what problems they can solve for people when they reach them through AI. You know, there are so many different tactics that are emerging for marketers to leverage in AI. Uh, of course, everyone's trying to [00:30:00] optimize so that they appear organically when users come and ask AI questions about their category, about their product, about the area that they play in.

Nate Elliott: Uh, of course, uh, we've had AI ads for a couple of years, uh, on smaller platforms, and ChatGPT's Still officially a beta, um, is moving forward, uh, aggressively, uh, every day. Uh, and we're starting to see brands integrate their own chatbots into tools like ChatGPT and Gemini. So these are all different ways that brands can potentially leverage AI.

Nate Elliott: But the thing they need to ask before they figure out which of those tactics makes the most sense is they need to understand why the audience they're targeting uses AI. Because if you don't know why someone's turning to these tools in the first place, how can you possibly help them get where they're going?

Nate Elliott: How can you possibly add value to that interaction? Mm-hmm. And one of the things that I learned all the way back in college when I was studying marketing was that you need to understand the [00:31:00] motivations of the people you're talking to so you can try to solve their problems. Mm-hmm. And if your product and your brand solves their problems, then they're a lot more likely to buy from you and to appreciate you.

Nate Elliott: If we can figure out as brands why people are turning to AI, what the motivations are that brought them to these tools in the first place, we have a lot better chance of tapping into those desires and motivations in a way that both suits the consumer's needs and helps us as companies as well.

Marcus Johnson: Mm-hmm.

Marcus Johnson: Perfect note to end on. Uh, I wish I could keep going with you guys, that's all we've got time for. The full report is called How People Use AI in 2026, our AI consumer survey reveals the building blocks of AI adoption. Link in the show notes, or you can head to pr- uh, EMARKETER.com if you're a Pro Plus subscriber.

Marcus Johnson: Um, that's all we've got time for for today. Thank you so much to my guests. Thank you first to Stephanie.

Stephanie Paterik: Thank you so much. Pleasure. This wasn't as hard as I was told that this is, like, part of EMARKETER hazing, but this was pleasant. This was downright pleasant. I told you that. I'm, I wo- I, uh, secret source.

Marcus Johnson: Thank you so much to [00:32:00] Nate. Nate. No problem.

Nate Elliott: Always a pleasure. Thanks, guys. Stephanie, I, I didn't tell you the podcast was the hazing. I, I told you that talking to Marcus was the hazing.

Marcus Johnson: Okay. Thanks, guys. Uh, thank you to the production crew. We've got Danny helping us out with this one, Danny runs our video.

Marcus Johnson: And to everyone for listening to Point of Numbers, uh, EMARKETER podcast made possible by AWIN. Thank you to you