Latest / The Adults In The AI Room / Mike Kaput On How To Get Value From Out-Of-The-Box AI Applications - Episode 2

Transcript

Mike Kaput On How To Get Value From Out-Of-The-Box AI Applications - Episode 2 · The Adults In The AI Room

• 0:07
  SPEAKER_00: Here we deal with leaders and practitioners and critical with AI in the enterprise as we move from the current state into the future of cognitive assistance and embedded AI in all aspects of our personal and professional lives.
• 0:22
  SPEAKER_00: Thanks so much for listening.
• 0:23
  SPEAKER_00: If this is your first time, we invite you to subscribe wherever you listen to podcasts.
• 0:28
  SPEAKER_00: And now here are your hosts, Seth Early and Chris Featherstone.
• 0:36
  SPEAKER_01: Hi, I'm Seth Early, and I'm here with my co-host, Chris Featherstone.
• 0:40
  SPEAKER_02: Seth, always a pleasure to uh to be with you.
• 0:44
  SPEAKER_02: And um these topics are super fun to discuss.
• 0:46
  SPEAKER_02: Um, one thing I do need to add is that all these opinions are my own and your own, and they don't affiliate with anything, you know, outside of this in the comedies.
• 0:55
  SPEAKER_02: That's that's the one disclaimer.
• 0:56
  SPEAKER_02: We'll put that in the beginning next time.
• 0:59
  SPEAKER_02: But it's it's it's fun to actually be with these folks, you know, like that we're gonna be talking with in just a sec.
• 1:05
  SPEAKER_02: And I think the key here is that it's one thing to talk about AI, and you know, a lot of a lot of people believe it's this mysticism that that needs to that you know needs to be in an application and that there's only a few select group of people that actually can implement it.
• 1:19
  SPEAKER_02: However, that's not the case.
• 1:20
  SPEAKER_02: So we're gonna demystify these things and get to the heart and soul of how now AI is table stakes in the application space, and low code or no code type scenarios are now super important for to enable people and business stakeholders and management and everything else to implement these things.
• 1:41
  SPEAKER_02: But I think to get at the heart and soul of testing these hypotheses.
• 1:45
  SPEAKER_02: Um, in fact, my my son, you know, he uh he got kind of a bad grade in his biology class and he looked at me and he's like, Dad, I will never use this in my life.
• 1:53
  SPEAKER_02: And I'm like, wait, what's what, science?
• 1:54
  SPEAKER_02: Are you kidding me?
• 1:56
  SPEAKER_02: And so now every chance we get, you know, my wife is like, Hey, did you see that game that you're playing?
• 2:00
  SPEAKER_02: Yeah, that was done because of science.
• 2:02
  SPEAKER_02: Hey, did you see that that food that you're eating?
• 2:04
  SPEAKER_02: That's done because of science.
• 2:05
  SPEAKER_02: Maybe the food's not a good example because it's designed, but but we're in this now, this space that every time we actually look at something, it's really really healthy to actually set that initial hypothesis regardless of the application.
• 2:19
  SPEAKER_02: And I think that's where uh we're gonna actually get some really, really great information, you know, today in terms of how to apply that in the application space.
• 2:29
  SPEAKER_02: Yeah.
• 2:29
  SPEAKER_01: So today uh yeah, and today we're going to be talking about uh how we can have AI tools that are increasingly embedded in various applications that don't require a lot of coding.
• 2:41
  SPEAKER_01: It doesn't require machine learning expertise, you don't need to be a data scientist or a data engineer.
• 2:46
  SPEAKER_01: They're really out-of-the-box types of applications.
• 2:50
  SPEAKER_01: And so our guest today is a man who has a real passion for making AI approachable, actionable, and accessible.
• 2:56
  SPEAKER_01: Uh, he does this as the chief content officer at Marketing AI Institute.
• 3:01
  SPEAKER_01: And the Marketing AI Institute provides online education for a worldwide audience of marketers and produces a yearly conference called the Marketing AI Conference, uh, which just wrapped up, as a matter of fact.
• 3:12
  SPEAKER_01: It's a 2021 event.
• 3:13
  SPEAKER_01: The next is going to be on the calendar for August of 2022.
• 3:17
  SPEAKER_01: And you can learn about that by going to uh ma i con.ai.
• 3:22
  SPEAKER_01: We'll have a uh link in the show notes uh for that.
• 3:25
  SPEAKER_01: With like with that, I'd like to welcome uh Mike Kaput to the show.
• 3:28
  SPEAKER_01: Welcome, Mike.
• 3:30
  SPEAKER_04: Seth, Chris, thanks so much for having me.
• 3:32
  SPEAKER_04: I'm excited.
• 3:33
  SPEAKER_04: That's awesome.
• 3:34
  SPEAKER_01: So go ahead.
• 3:35
  SPEAKER_01: Uh Chris, do you want to go ahead and uh uh kick us off in terms of getting started with uh some of our uh discussion with Mike?
• 3:42
  SPEAKER_02: Yeah, absolutely.
• 3:42
  SPEAKER_02: So, Mike, thanks, thanks again for uh for joining us.
• 3:45
  SPEAKER_02: I think what would be super interesting is first, you you know, why don't you give us just a simple overview of your day-to-day?
• 3:51
  SPEAKER_02: You know, what you what you're doing, what you're working on, you know, all the good stuff that that the these folks out out in the uh in the internet need to hear.
• 3:59
  SPEAKER_04: Yeah, that's awesome.
• 4:00
  SPEAKER_04: That's a great question.
• 4:01
  SPEAKER_04: Um, so yeah, my day to day is really concerned with uh building an audience for marketing AI Institute.
• 4:08
  SPEAKER_04: So we're trying to reach as many marketing professionals as possible through our written audio and visual content.
• 4:15
  SPEAKER_04: So in any given day, I'm either creating content, promoting content, or figuring out how to make our content strategy work better.
• 4:23
  SPEAKER_04: And as a part of that, taking all the traffic we get or are trying to get from our site and turn them into an engaged audience.
• 4:32
  SPEAKER_04: And there's a bit of a community element to that as well.
• 4:36
  SPEAKER_04: And there's also a serious um kind of revenue operations element as well as they move further down the funnel and become paying members of our online education products, our event, um, or just uh as part of our community.
• 4:51
  SPEAKER_04: So on any given day, I'm trying to increase traffic, increase leads, and hopefully convert some leads into uh happy customers.
• 4:59
  SPEAKER_01: And by the way, sorry, I should mention how Mike and I met.
• 5:03
  SPEAKER_01: Uh, when I was doing research for my book, The AI Powered Enterprise, um, I was looking at different uh events and venues, and I came across this marketing AI uh event.
• 5:13
  SPEAKER_01: And I thought, you know, I had low expectations.
• 5:16
  SPEAKER_01: I was like, okay, these are going to be marketers, this is gonna be arm waving, this is gonna be a bunch of, you know, blah, blah, blah, blah, you know, marketing crap.
• 5:23
  SPEAKER_01: Uh, and I was very surprised to see the depth and detail and the quality of the presentations.
• 5:29
  SPEAKER_01: Actually, a number of my case studies for the book uh came directly out of that conference.
• 5:34
  SPEAKER_01: So it's a really high quality event and it's uh I highly recommend it.
• 5:37
  SPEAKER_01: So I just wanted to put that uh little bit of context in into the uh discussion.
• 5:42
  SPEAKER_02: I'd love to get an idea too, Mike, of you know, where you started, because I think it's fascinating that you started as a journalist in Egypt.
• 5:49
  SPEAKER_02: And that's generally, you know, not the course I took or uh that I know Seth took.
• 5:54
  SPEAKER_02: So give us some details on just some of your background because that's it's fascinating.
• 5:58
  SPEAKER_04: Yeah, sure.
• 5:59
  SPEAKER_04: Um, so I started out um as basically a political science major in college, thought I probably wanted to do something in government and ended up studying abroad in Cairo during my school years and then moving back there, um, kind of in like the depths of the economic crisis, kind of didn't know what I wanted to do.
• 6:20
  SPEAKER_04: And, you know, studying Arabic in Egypt again seemed like a good idea if I was gonna end up, you know, going the government route.
• 6:28
  SPEAKER_04: Um, you know, once I got there, my plans changed pretty seriously, but I ended up falling into uh working for a couple of magazines, doing both reporting and editing, and just really loved it.
• 6:40
  SPEAKER_04: I'd always been someone that loved writing, reading, and just learning about the world.
• 6:44
  SPEAKER_04: So it was a really good fit at the time.
• 6:47
  SPEAKER_04: Um, worked in that for a few years, ended up moving back to the States where I was doing kind of my own freelance writing, marketing, consulting type business, but still mostly focused on creating traditional content.
• 7:02
  SPEAKER_04: Um and, you know, as anyone who has read the news or not bought the news recently, the world of journalism kind of has been turned upside down over the uh last 10 to 15 years.
• 7:15
  SPEAKER_04: And I kind of saw that and realized, you know, after seeing colleagues, friends laid off all the time and just kind of overworked and underpaid, I was like, I think I think there may be some other ways we could use these skills and kind of pivot.
• 7:30
  SPEAKER_04: So I that's how I ended up uh working for a marketing agency.
• 7:34
  SPEAKER_04: And that agency is called PR 2020.
• 7:36
  SPEAKER_04: And I'm still uh still work with them today because PR 2020, um, our CEO and founder, Paul Reitzer, um, has always been a person that has been focused on marketing automation, where technology is going in the marketing industry.
• 7:53
  SPEAKER_04: We were HubSpot's first ever agency partner.
• 7:57
  SPEAKER_04: So I got my start cutting my teeth on using content marketing and marketing automation to grow clients' businesses.
• 8:05
  SPEAKER_04: And as part of that, myself and Paul kind of got geeked out and interested in artificial intelligence.
• 8:13
  SPEAKER_04: Uh, Paul, for him, it was many, many years back.
• 8:16
  SPEAKER_04: Uh, for me, it was probably about six years ago.
• 8:19
  SPEAKER_04: We really started seriously talking about it.
• 8:21
  SPEAKER_04: Um, at that time, we had just heard about AI, we had read a couple articles, maybe a book or two, and we were starting to think, okay, maybe this could have a real impact on our business, our clients' businesses, and um, and maybe drive growth in a way that we typically hadn't had access to.
• 8:42
  SPEAKER_04: I mean, we were still very deep into kind of the inbound marketing revolution, the marketing automation revolution, just understanding how technology was fundamentally changing marketing.
• 8:53
  SPEAKER_04: Um, we really were at the forefront of living through the challenges and opportunities presented by that.
• 8:59
  SPEAKER_04: So we started just kind of writing um some articles about what we were learning, started sharing our knowledge as almost a side project to start, just to see if there was any other marketing uh professionals out there who were struggling with some of the same things we were.
• 9:17
  SPEAKER_04: We were trying to understand AI, we were trying to find tools.
• 9:22
  SPEAKER_04: We were already successful and you know, smart technology um purchasers through the agency.
• 9:29
  SPEAKER_04: You know, we knew certain questions to ask, but we kept running up against this idea that it just seemed like there was a lot of confusion in the market.
• 9:37
  SPEAKER_04: It seemed like there was a lot of buzzwords, it seemed like there was a lack of connecting the dots between this kind of interesting technology that is often overhyped, but is it actually very powerful?
• 9:49
  SPEAKER_04: And actually using it for business results.
• 9:52
  SPEAKER_04: So that's where we ended up gravitating towards and we started building a very large audience for what we were writing and talking about.
• 10:01
  SPEAKER_04: So, long story short, a couple of years ago, Marketing AI Institute spun off into its own company, and we've just tried to really accelerate the growth of it from there.
• 10:11
  SPEAKER_04: We launched our conference in 19.
• 10:13
  SPEAKER_04: Um, we launched AI Academy for Marketers in 2020, kind of as a response.
• 10:18
  SPEAKER_04: Unfortunately, we had to cancel our event and move up the plans for that.
• 10:21
  SPEAKER_04: And we also then took on um about a million or so dollars in seed money um late last year because we realized that uh, you know, between COVID, between the trends we were seeing and between what we had observed over the last 15 years in the marketing industry, uh, we made a big bet.
• 10:39
  SPEAKER_04: And I think it's the right bet that AI's time has come.
• 10:43
  SPEAKER_02: That's awesome.
• 10:43
  SPEAKER_02: That's awesome.
• 10:44
  SPEAKER_02: There's, I mean, most people in you know, in my experience, they look at you know the marketing side and they think, well, I'm automatically gonna just look at this as how do we do more personalization?
• 10:54
  SPEAKER_02: You know what I mean, in a commerce setting, which you know, let's shelf that for a while because the beauty of what we're talking about here with the content is super, super critical.
• 11:05
  SPEAKER_02: And it's not just what the content is, but how to place the content appropriately, what needs to be developed, what needs what's there, and all that kind of stuff.
• 11:12
  SPEAKER_02: So I'd love to get a sense, you know, to to as we you know double click into this notion of you being the con, you know, the chief content officer and then applying AI to the content.
• 11:25
  SPEAKER_02: What are the what are like maybe your top use cases that you're that you're driving right now or that you're putting into?
• 11:31
  SPEAKER_04: That's a really good question.
• 11:32
  SPEAKER_04: So a couple of the biggest things that we today use AI for, and I should include the caveat that you know we're always growing and learning.
• 11:40
  SPEAKER_04: So some of these tools, you know, we're experimenting.
• 11:43
  SPEAKER_04: I'm using at any given time anywhere from half a dozen to a dozen different AI tools and platforms and kind of kicking the tires on them.
• 11:50
  SPEAKER_04: So we definitely have a few tools that we always recommend that we we are actively using really in depth.
• 11:57
  SPEAKER_04: And then we've got many others that we're just kind of trying out and seeing how they work, what they can do.
• 12:03
  SPEAKER_04: So just so people know, you know, we're always we're always looking to apply this to something else.
• 12:08
  SPEAKER_04: So if you have any positive, negative experiences with certain use cases or vendors or tools or things like that, we always are interested in hearing them.
• 12:18
  SPEAKER_04: But from our perspective, what we're really doing a lot with when it comes to AI is for lack of uh if I had to nest all these use cases under one thing, it's scaling content marketing and scaling content programs.
• 12:33
  SPEAKER_04: Because at the end of the day, one of the biggest things we started looking into that got us really interested in AI in the first place was hearing stories about how certain AI systems supposedly could write news articles.
• 12:46
  SPEAKER_04: And this is an interesting tie back to what I used to do.
• 12:50
  SPEAKER_04: It's like, oh, okay, a machine can do that.
• 12:52
  SPEAKER_04: That's really interesting.
• 12:52
  SPEAKER_04: We were seeing like uh Associate Press had used some type of tools to automate writing earnings reports way back when for like stock earnings at companies.
• 13:03
  SPEAKER_04: And that kind of led us down this rabbit hole of saying, like, you know, we had at the time at our agency had been hiring a lot of writers.
• 13:12
  SPEAKER_04: We looked for people with journalism backgrounds, which is how I got my start, because we were doing so much content marketing for clients.
• 13:19
  SPEAKER_04: The problem is, as anyone who has created content knows, it takes a really long time and it doesn't always work.
• 13:25
  SPEAKER_04: So our big thing is looking for AI use cases, at least in my day-to-day.
• 13:30
  SPEAKER_04: And by no means do we only use AI for content marketing.
• 13:33
  SPEAKER_04: So we can definitely talk about some other use cases there.
• 13:36
  SPEAKER_04: But in the content space, we're really looking at tools that can help us identify what type of content we should be creating, help us brainstorm what should be included in the pieces of content we're creating in order to either rank in search or appeal to the intent behind what people are looking for.
• 13:56
  SPEAKER_04: We also are using certain tools to actually create parts of the content itself.
• 14:01
  SPEAKER_04: So a lot of that content creation use case right now is with written content because language models have actually dramatically improved in probably the last year, which is one of the really interesting signals that we're finally coming into this area of like, oh my God, AI is going to start really changing how people do their jobs, I think.
• 14:23
  SPEAKER_04: And then, you know, we're doing things like looking at other ways AI can actually help us scale the creation of content.
• 14:30
  SPEAKER_04: And that could be as simple as uh telling us what to write so we do know we do have confidence we can go hire someone to do it.
• 14:38
  SPEAKER_04: We do have confidence we could spend budget on freelancers to create this content rather than just kind of spraying and praying a bunch of content and seeing, you know, just trusting luck.
• 14:49
  SPEAKER_04: It also helps us get better and understand what content are people actually interested in and how can we give them more of the topics that they actually care about.
• 14:59
  SPEAKER_04: So there's everything, there is a heavy element of content creation, content strategy, and personalization in what we do with AI today.
• 15:07
  SPEAKER_02: One one quick thought, Seth.
• 15:09
  SPEAKER_02: Sorry, I was I was just thinking about this because I know you want to jump into some of the you know the deeper areas.
• 15:13
  SPEAKER_02: What about user-driven content?
• 15:15
  SPEAKER_02: Where do you see user-driven content you know play a role in this?
• 15:18
  SPEAKER_04: That's a really good question.
• 15:20
  SPEAKER_04: I think that's probably one of the areas we haven't really gotten to thinking about yet, because one of the areas that we really do want to scale is having more uh user-generated content actually uh powering our site.
• 15:32
  SPEAKER_04: I mean, the dream here for us, and you know, don't quote me on this, but if we do our job right, we really do want to build through the institute's blog, through the website, through our podcast, webinars, et cetera.
• 15:46
  SPEAKER_04: We want to build from the ground up the AI-powered content property of the future, not just apply AI to use cases, which we're doing now and is producing good results, but really rethink from the ground up what is an AI-first content property look like?
• 16:02
  SPEAKER_04: Because I'm not sure anyone has done it yet.
• 16:05
  SPEAKER_04: There's people doing probably doing things like we're doing or doing better than we are, but we're really going to hopefully reinvent the wheel as we go here.
• 16:14
  SPEAKER_04: So I think user-generated content is a huge piece of that.
• 16:17
  SPEAKER_04: We haven't gotten to it yet.
• 16:19
  SPEAKER_01: So one of the things that we were talking about in terms of this description of the session is that, you know, a lot of people think of AI initiatives as requiring a lot of heavy lifting, a lot of preparation, a lot of work on data sources.
• 16:32
  SPEAKER_01: You know, there's just these, you know, lots of barriers and lots of things that need to be done in preparation.
• 16:37
  SPEAKER_01: But you know, the premise here was, you know, there's stuff that we can do that's kind of out of the box or that's baked into other technologies that we really don't have to do a lot of that kind of heavy lifting for.
• 16:48
  SPEAKER_01: Um, so so when we talk about out of the box, how would you define that kind of AI flavor tech uh tech that flavor of AI technology?
• 16:58
  SPEAKER_01: And what really is out of the box versus what do you have to do to make that stuff work?
• 17:03
  SPEAKER_04: That's a really good question.
• 17:04
  SPEAKER_04: And it's a huge challenge because to your point, there's so many solutions that may be mentioned and may work really well, but are not suitable for, say, a smaller business or a business without the right foundation.
• 17:17
  SPEAKER_04: So the good news for us is that we are actively a small startup.
• 17:22
  SPEAKER_04: So we are forced to find things that work out of the box on a relatively uh small budget.
• 17:28
  SPEAKER_04: So a lot of the things I'm gonna mention, not every single one I mentioned today, is probably gonna be strictly immediately out of the box, but they are things that we have used and we're able to use with a relatively small team.
• 17:41
  SPEAKER_04: Um, and I think when we think about out of the box, what we really mean is can your average marketer get started with it without having to consult any type of super technical lead on your team?
• 17:55
  SPEAKER_04: Now, I think that should always be happening where you are consulting technical experts for your strategy, for deeper implementations, but some of these tools really function very much like your average SaaS tool, where you can just sign up for a free trial, pull out a credit card, and start using them.
• 18:12
  SPEAKER_04: And that's kind of to me the most exciting out of the box things is the ones that, and I literally have done this where even if someone's not in the office or I can't find someone with uh, you know, a company credit card, or I leave mine at home, like I just whip out my card and start paying for a month of these tools and start using them.
• 18:33
  SPEAKER_04: And that's really exciting to me is I don't think we were able to do that even probably three years ago for most of this.
• 18:39
  SPEAKER_01: And a follow-up question to that.
• 18:40
  SPEAKER_01: So that's valuable for experimentation, uh, but you still need to have supporting processes, you still need to understand your customer engagement strategy, you still need to do some work around, you know, what content is strategically important.
• 18:54
  SPEAKER_01: I mean, one of the tools that I wrote about in the book was market views and the fact that there's a great way of kind of looking at the semantic space that a piece of that a concept will be around and then building out related content.
• 19:06
  SPEAKER_01: So building that content hub and then, you know, understand the terminology your customers are using.
• 19:11
  SPEAKER_01: But the value of that was being able to uh kind of build that framework for that.
• 19:17
  SPEAKER_01: But you still had to understand what's your, you know, content strategy at a foundational level and what terminology is important, what concepts and so on.
• 19:25
  SPEAKER_01: So, do you want to talk a little bit about kind of the supporting processes that need to be part of this, even though it's quote out of the box from a technical perspective or a data perspective, and you don't need IT, you still need to think differently about, you know, what's the same about the marketer's job and what's different when it comes to some of these out of the box tools?
• 19:44
  SPEAKER_04: Yeah, that's a great question.
• 19:46
  SPEAKER_04: I think I joke that if you scratch an itch for AI, you pretty quickly just run into like a typical marketing strategy problem because a lot of the things people sometimes sit down and say, Well, I've heard so much about AI.
• 19:59
  SPEAKER_04: Can't we just use AI to solve this problem?
• 20:01
  SPEAKER_04: And the answer is probably yes, but but it needs so much foundational work in terms of your processes.
• 20:09
  SPEAKER_04: So honestly, one of the things we have learned the hard way, because we've uh swung and missed on a couple pilots here that really just set us back in terms of time and money, is that I would not be uh trying to adopt seriously any AI tool without your basic marketing strategy and processes documented all together.
• 20:31
  SPEAKER_04: You might call it almost like a if you think about more what some of the like marketing ops people do when they're documenting, like what does our tech stack look like?
• 20:39
  SPEAKER_04: What do our processes and approvals look like?
• 20:41
  SPEAKER_04: I truly think it sounds so mundane, but it is so important to say, okay, what is the customer journey like?
• 20:47
  SPEAKER_04: What does your strategy look like?
• 20:48
  SPEAKER_04: Even if it's not perfect, right?
• 20:50
  SPEAKER_04: It needs to be centralized in one place, well documented, maybe visual would be even helpful.
• 20:57
  SPEAKER_04: And then you can say, okay, where does AI plug and play into these processes?
• 21:02
  SPEAKER_01: I have a little phrase I like to say.
• 21:05
  SPEAKER_01: Uh I say you can't automate a mess.
• 21:08
  SPEAKER_01: And you can't automate what you don't understand, right?
• 21:11
  SPEAKER_01: So if you don't understand it, the AI tool is not going to figure out for you.
• 21:15
  SPEAKER_01: Humans need to be able to do that, understand the strategy, understand the process.
• 21:19
  SPEAKER_01: And then you're really looking for interventions in steps of that process as opposed to some big grand galactic Uber, you know, AI that's going to solve all your problems.
• 21:28
  SPEAKER_01: Um, so so the the upside of this is you get to experiment, you get to try and fail if you, you know, and and it at low cost, at low risk.
• 21:36
  SPEAKER_01: Uh, what is the downside when you start to mention you can do all these?
• 21:40
  SPEAKER_01: It kind of becomes a little shadow IT y, right?
• 21:43
  SPEAKER_01: And you can do all these independent things.
• 21:45
  SPEAKER_01: Well, what happens when you get these experiments?
• 21:48
  SPEAKER_01: What do you need to do?
• 21:49
  SPEAKER_01: What do you need to be aware of or watch out for when you're doing these experiments and then trying to bring them together or trying to centralize?
• 21:56
  SPEAKER_01: What do you need to be aware of in terms of standards or processes?
• 22:00
  SPEAKER_01: or methodologies and so on.
• 22:01
  SPEAKER_01: So you don't end up with more fragmentation and more, you know, technologies and so on that, you know, people want to pull their hair off.
• 22:10
  SPEAKER_04: Yeah, that I and that's always that, always that danger, right?
• 22:14
  SPEAKER_04: Is the more we try to adopt a certain technology, the more complex we're going to be like technology is going to create a problem all over again that we need to solve.
• 22:23
  SPEAKER_04: But I think my read on the marketing AI specifically market um is that it is still early enough.
• 22:32
  SPEAKER_04: I think fragmentation unfortunately is going to just be a cost of doing business for the time being.
• 22:38
  SPEAKER_04: I don't think that will always be the case.
• 22:40
  SPEAKER_04: But one big thing we're seeing is that, you know, we do anticipate at some point some of the bigger platforms you're already using, like a HubSpot, they have some AI features that are interesting you're probably already using and not knowing it.
• 22:54
  SPEAKER_04: But we would fully anticipate down the line those types of companies having all of these capabilities in one platform.
• 23:02
  SPEAKER_04: But right now that's not the case.
• 23:03
  SPEAKER_04: So what becomes critical is sitting down before you even talk about or read about AI is sitting down and understanding at a deep level your use cases and really taking the time to sit down and say, okay, what are the business problems we are trying to solve?
• 23:21
  SPEAKER_04: What do these look like in their smallest, most granular form in terms of, and it can be, you know, I say our problem is scaling content marketing, but that's a that's an enormous problem.
• 23:31
  SPEAKER_04: That really boils down into about 30 or 40 different things that we need to do better.
• 23:37
  SPEAKER_04: So we may start extremely small and say, you know, it's kind of a real pain that every time we write a blog post, we have to go do this keyword research.
• 23:48
  SPEAKER_04: We don't have visibility into the intent of the search.
• 23:51
  SPEAKER_04: We don't have even, you know, something more than generalized search data, is there a smarter way that we could pick the topics to write about that's really that core problem, that use case where you say, okay, now I'm going to go say is there AI that can help me with content topics or with SEO research or keyword research.
• 24:12
  SPEAKER_04: So I think you have to be really careful upfront and saying here's the handful of very specific things I want to achieve.
• 24:20
  SPEAKER_04: I don't know if AI can achieve them, but I'm going to go look if it can.
• 24:24
  SPEAKER_04: I'm going to go start doing research and talking to people, listening to podcasts like this, and see if I can figure out AI for my specific use case.
• 24:33
  SPEAKER_04: So that's critical in my opinion.
• 24:36
  SPEAKER_01: That's a great way to think about it.
• 24:37
  SPEAKER_01: And we do talk a lot about you know really defining those individuals, you know, specific things that where you can have an intervention and impact.
• 24:45
  SPEAKER_01: I do want to kind of throw a question to you Chris and to both of you really.
• 24:49
  SPEAKER_01: And and you know one of the things that Mike was just mentioning is that you know the hub spots of the world and first of all search has been using machine learning at one level in text analytics for a long time and that's a flavor of AI.
• 25:01
  SPEAKER_01: And these tools, these platforms are increasingly do it using it behind the scenes.
• 25:06
  SPEAKER_01: But there's so much change and evolution and and pace, fast pace of uh these things changing that you know you need kind of best of breed.
• 25:16
  SPEAKER_01: They're never going to be able to do everything for everybody.
• 25:18
  SPEAKER_01: And and you really do need to think about so Chris in your experience you know how do you think about kind of switching in and out pieces to provide very specific functionality and then how do you kind of you know manage that longer term you've been in this space for quite a while.
• 25:33
  SPEAKER_01: So what are your thoughts on that?
• 25:34
  SPEAKER_02: So it's just definitely it's it's it's it has to go back to the use cases and like like you know Mike had talked about the simple hypotheses of what needs to be asked first and then it's an evolution.
• 25:47
  SPEAKER_02: The best um you know environments out there should have you know reinforcement learning models that can evolve with these types of questions, right?
• 25:56
  SPEAKER_02: Or with these types of of ideas.
• 25:58
  SPEAKER_02: So if we do it especially across language speech and vision, right, we're going to train and tune these models to actually figure out what information we need.
• 26:07
  SPEAKER_02: So if it's a speech recognition model, it's all about accuracy.
• 26:11
  SPEAKER_02: But that accuracy may be semantic according to what's being spoken because depending on the pharmacy is going to be different than legal and medical oh yeah absolutely so those those key terms phrases and semantic terminology have to be tuned and trained in the model so that you know the accuracy is super high so that you can actually do something with it.
• 26:34
  SPEAKER_02: I mean like the the crazy fallacy about artificial intelligence is not that we can train and tune these models to generate stuff.
• 26:42
  SPEAKER_02: It's the recommendation engines and what we do with it after the fact.
• 26:46
  SPEAKER_02: Because we're actually you know applying all these machines to go say hey I want to understand you know why Chris's hairline is receding right and is there a product that can map to that and and the answer will be here's all the data that shows this now what do we do with it?
• 27:03
  SPEAKER_02: And that's the that's where it gets into you know whether it's a pre-built model or it's a a completely custom trained and tuned model, you know, something that's out of the box or something that's super you know that we have to you know build ourselves a data scientist and stuff like that.
• 27:18
  SPEAKER_02: There's a lot of dependencies but it all goes back to asking that that simple question of you know of what is it that I'm trying to look for and what am I going to do with it.
• 27:29
  unknown: Yeah.
• 27:29
  SPEAKER_02: Right?
• 27:29
  SPEAKER_01: I always like to say the first thing about AI is forget about AI.
• 27:33
  SPEAKER_02: Yeah yeah right exactly they're data generators yeah yeah yeah yeah I you know I'll tell you what one thing that um that came to mind too Mike as you were discussing this because we have you know a bunch of different use cases and things like that I want to get to at some point where you're in your opinion where somebody should start but before that I wanted to ask you where does ethics play a role in this yeah that's I feel like that is a probably a trillion dollar question on a long enough timeline here because um I don't know what your experience has been, but I think a lot of people have that question but not enough people are probably talking about it in a robust way like I feel like everyone knows it's important, right?
• 28:17
  SPEAKER_04: That these systems can go wrong.
• 28:19
  SPEAKER_04: There are ethical concerns there are bias concerns.
• 28:22
  SPEAKER_04: I don't think I think few people I've seen so far have really been diving deep into these issues.
• 28:29
  SPEAKER_04: And for brands it's going to be critical because the moment you have an here's like a good practical example that I wrote about a few years ago it you know it didn't happen super recently but like uh Apple released like a credit card and they got put on blast by an entrepreneur who had like millions of Twitter followers because him and his wife who's also an entrepreneur they had like the same financial picture they both applied for this card and it was all algorithmic how you got credit approvals and he got approved for 10 times the amount of credit that his wife did.
• 29:07
  SPEAKER_04: Now the interesting thing is so he was furious and he was like what is this this seems like a other I don't know why that is but right right yeah yeah I was like I wish I had this problem but um uh but it was one of these things where he is basically doing this in real time the value of the algorithm is hey we can automatically approve people for credit we don't have to go through a lengthy process.
• 29:33
  SPEAKER_04: The downside is that this happens so fast it happens in real time within 30 minutes Apple had a massive PR issue on their hands.
• 29:42
  SPEAKER_04: And the worst part is as they tried to navigate this issue in real time, they couldn't answer why the algorithm had done it because they didn't build it.
• 29:52
  SPEAKER_04: Goldman Sachs ended up building it.
• 29:54
  SPEAKER_04: Goldman Sachs wasn't really available to talk about it.
• 29:57
  SPEAKER_04: So suddenly you have this explainability issue where where it's having a real effect on someone's life, which is the main thing is like you never want something that's discriminating against someone because they're a woman.
• 30:09
  SPEAKER_04: But also you have this second order effect sort of thing where Apple is losing uh brand equity, dealing with PR nightmares and then this third order thing where nobody knows what who's actually responsible for this.
• 30:24
  SPEAKER_04: Like Apple literally said hey sorry we we can't really help you.
• 30:27
  SPEAKER_04: We didn't make it so it's this that alone is just one simple example of all these considerations around ethics bias and explainability that nobody is really really sitting down and building a strategy around there should be a crisis communication strategy that's like kind of our world where we come from just for AI tools.
• 30:47
  SPEAKER_01: Like what happens when they go wrong once you've started adopting them well I following question that is what really should what are the biggest risks to a marketer and where should they keep their you know their radar up around those?
• 30:59
  SPEAKER_04: So what you know what should they be looking for out of box tools especially when you don't have explainability or there could be a third party or you know some other tertiary thing that that people don't know about you know how do they safeguard their brand and how do they kind of you know be aware of this what questions should they ask I mean that seems like a very complex area but important yeah I think I think the first question I would probably be asking of any vendor is where is the data coming from how is the machine making its predictions now they're not going to either want to or be able to tell you every nuance of it but no but knowing like okay where are the inherent biases in this data and keep in mind when we say that we're not necessarily saying just bias in the sense that an algorithm discriminated against someone though that is kind of a blatant example it can also just be a biased data set not against any group of people but just optimized for the wrong outcomes right so if you have say you know it's like any survey on earth if someone says hey we surveyed uh 4,000 business professionals but they all were in the same industry you'd be like okay I don't know if this is a widely applicable right survey right so you have to watch out for both of those and understand where is the data coming from.
• 32:15
  SPEAKER_04: Honestly, I think it makes sense to get creative and start brainstorming like what is the worst case scenario here?
• 32:21
  SPEAKER_04: Where does this go wrong?
• 32:23
  SPEAKER_04: And I think you understand that by saying to a vendor or understanding as you demo, what is the machine doing versus what is the human doing?
• 32:31
  SPEAKER_04: Because then you can really isolate like how much control do we have over this?
• 32:35
  SPEAKER_04: And I don't mean to scare anyone with that because a lot of these tools that we're talking about using for marketing, there's no way it's not like running off and talking to your customers necessarily on its own with anything you have not approved.
• 32:48
  SPEAKER_04: However, when you do get into these more sophisticated recommendations and personalization things, then it can get a little interesting, I think.
• 32:57
  SPEAKER_02: Yeah well you you you nailed it where it's it's all the desired outcome, right?
• 33:02
  SPEAKER_02: Because there is some scary stuff out in social media world where it is a desired outcome of what the algorithm should be doing to increase more um attention, right?
• 33:12
  SPEAKER_02: Those kind of things.
• 33:13
  SPEAKER_02: But I think you're spot on where if you're an organization, you know, it's not necessarily like you said, you know, it's all cast in jello right now.
• 33:22
  SPEAKER_02: There's nothing super concrete about how to apply ethics.
• 33:26
  SPEAKER_02: So and it's kind of like the parallel of what we used to do with security where we used to just ask and then create like a security profile.
• 33:34
  SPEAKER_02: Like what should we be thinking about and just like ask some simple questions just to make sure that we're compliant with whatever those needs are and I think from an ethics perspective I think you're spot on where it's just a matter of hey let's just create an ethics profile for what this could look like and just create a contrary view to the derived or the desired outcomes that we're testing for, knowing that we have good data, bad data, whatever it is, right?
• 34:02
  SPEAKER_02: We want to test against the most um the most you know the best data we have in order to drive that outcome.
• 34:08
  SPEAKER_02: But I think it's just a matter of just asking yourself and putting together just a simple ethics profile to ask ask just some simple questions and say okay cool do we meet these standards and of that it's just like what we're doing with the hypotheses anyway we're gonna come out with other outcomes and say oh we didn't think about that because we didn't know to think about that and we're gonna miss 100% of the things we aren't thinking about.
• 34:28
  SPEAKER_02: So let's you know let's be open to some some different things.
• 34:32
  SPEAKER_01: So I think what's the worst what's the downside?
• 34:34
  SPEAKER_01: What's the worst case scenario?
• 34:36
  SPEAKER_01: How could it impact?
• 34:37
  SPEAKER_01: And again I should say if there's human intervention and human judgment before things go out if it's not making decisions that are going to impact the brand not making recommendations that that you don't have some control over this lower risk.
• 34:49
  SPEAKER_01: You know when you start looking at um the stuff that's coming out from the Wall Street Journal about Facebook, I mean you can see that there's some really really thorny questions that they get into with those algorithms being optimized strictly for engagement, regardless of you know, which is high emotional content, right?
• 35:05
  SPEAKER_01: We don't want to get into the details of that necessarily but we but you know the observation from from the the reports were that you know they're aware of certain things and chose profit and engagement over you know safety and and moderation.
• 35:18
  SPEAKER_01: So I mean obviously that's a that's an extreme uh case of that but I think what you're what we're talking about is understanding what those risks could be and asking those questions to make sure you have control.
• 35:29
  SPEAKER_02: It it's the other way too though it's not just you know we don't want to harm but also the beauty of what some of these things can do is expose expose really interesting data data markers and connection points that we didn't see before like you know we have a a ton of these genealogy systems that are out there on the on the you know in the interwebs now and what they're doing though is they're associating now the genealogical profiles of people in different ways to not rewrite their history but to tell them the actual more more truthful history and some of these folks don't like it.
• 36:02
  SPEAKER_02: You know what I mean?
• 36:03
  SPEAKER_02: And it's I'm sorry but it's just the way that is but this is what we're also so the ethics are also coming in terms of how do we provide more truthful outcomes.
• 36:13
  SPEAKER_02: You know what I mean?
• 36:14
  SPEAKER_02: And then you're right it then is hopefully you don't get some um you know some uh negative intent or or you know all centered around doing something negative with those outcomes and with that data right and apply it somewhere so I want to switch gears just a bit um into more along lines of you know let's say that I'm um I'm just hearing for this you know not hearing this for the first time but I'm struggling or thinking about how to apply AI.
• 36:43
  SPEAKER_02: You know, in your opinion Mike where should one start right and knowing that hey listen I've got an organization I'm doing basic marketing today top of the funnel type of activities creating content I've got my creatives in a pretty you know good workflow of you know seeing the you know going through the creative breath all the way out to you know content publishing and things.
• 37:03
  SPEAKER_02: If I wanted to add some some of these things what use cases should I start with you know in terms of you know kind of a crawl rock run you know approach where would you where would you begin or what would you yeah that's it that's a great question.
• 37:15
  SPEAKER_04: I think it sounds simple but I think it's important.
• 37:19
  SPEAKER_04: I would sit down with a spreadsheet and write down everything you do in a day a week a month and a quarter and it can literally just be like write blog posts.
• 37:32
  SPEAKER_04: I write those every week cool okay I do performance reports for my boss every quarter or week or month or whatever.
• 37:42
  SPEAKER_04: I would do that for you personally to start.
• 37:45
  SPEAKER_04: And then I would if you have the insight in it if you're a smaller organization like do it for your team or have them spend 10 minutes doing it because then I think what you end up with pretty quickly is a list of all the activities that take up your time.
• 38:00
  SPEAKER_04: And I think the saving time is not the only opportunity with AI and probably and not the biggest one frankly but I think it is a good starting one because it's easy to understand where if you say hey I have this list of activities that I do all the time I would literally just go through and I kind of like this kind of subjective way I would think about this is which of these do I dread?
• 38:24
  SPEAKER_04: Like which of these are the ones I really don't like.
• 38:26
  SPEAKER_04: For me it's like just putting together performance reports that have stuff I already know in them, right?
• 38:31
  SPEAKER_04: So they take me forever it's really repetitive it still relies on data that's actually a really good starting use case for possibly using an AI tool to solve something and you're not taking away work or activities that either are giving you value or that you like doing you're just net positive in the sense of looking how to do something smarter that you hate doing.
• 38:59
  SPEAKER_04: And so I think starting those really narrow things where you say okay these are like the top three things that take up the most time the most energy the most money and that we all hate is there a way to do them in a smarter way and there that's when you end up saying okay that's actually a really good place to start looking into a small AI pilot for those activities.
• 39:22
  SPEAKER_01: So um I think it's a really great way to think about it, right?
• 39:25
  SPEAKER_01: Because when you when you look at what AI is doing and is going to continue to do it's going to reduce uh work that people you know it'll reduce a lot of the work that people don't want to do anyway right yeah like uh you know there used to be there's always been concerns about automation taking away jobs you know when when we first started using you know steam shovels at the you know turn of the 20th century uh you know what are all those ditch diggers going to do?
• 39:50
  SPEAKER_01: Well people don't like digging digging ditches right and and and what can we do?
• 39:54
  SPEAKER_01: We can build cities and and and uh skyscrapers and highways or whatever you know my point is that there's all sorts of things we could do once we automated some of those manual processes and it's kind of the same way but we're still going to have to you know reskill people and and uh you know give them different direction and so on um but when you think about um uh a lot of these tools that are going to be used to uh improve those processes those will provide some competitive advantage right reducing costs improving services when you think about out of the box you know and and what comes to mind are say personalization algorithms that that you know some vendors are saying well you can just bolt this on to your your shopping cart or your e-commerce site and it'll it'll it'll improve this these recommendations but if it's uh if it's kind of something that you can do out of the box well how much competitive advantage is it providing because it's it's raising the bar and you know you can get you know standardization is great for efficiency but differentiation is great for competitive advantage so where do you kind of draw the line in terms of okay we're going to do this out of the box thing but everybody else all of our competitors can do that and eventually they'll get there.
• 41:08
  SPEAKER_01: But then when do we need to kind of go a step further and think, well this is really to differentiate as opposed to standardize is that is that a fair question I think that's a that's probably another sort of billion or trillion dollar question honestly so I think that there's a couple threads to this.
• 41:25
  SPEAKER_04: So one I think there's a big opportunity because there are so many say at least in my space in marketing there's so many marketers who aren't using any of these tools and there's a couple uh sort of forward thinking brands that are and they're reaping massive competitive advantages because of it.
• 41:44
  SPEAKER_04: So on that from a competitive advantage standpoint today I don't think you can afford not to be using some of these tools if you're able to but as adoption increases it becomes more widespread there will probably be diminishing returns but I would expect probably the technology will get better as a result and will actually lift all boats.
• 42:09
  SPEAKER_04: But I also think long time from now or at least further down the line we need to free ourselves up to do those human things that are the differentiators like story is so important.
• 42:21
  SPEAKER_04: It always has been but telling the right story to your market, you probably need to be devoting a lot more time to that than you do today.
• 42:30
  SPEAKER_04: And that is going to be something you will hopefully are enabled to do more of by adopting some of these tools.
• 42:37
  SPEAKER_04: But I think it'll it's a fast moving landscape.
• 42:39
  SPEAKER_04: I mean like I was saying with the um I I'm pretty surprised actually language models have gotten to where they are now as fast as they did.
• 42:47
  SPEAKER_04: And most of that progress probably happened in the last 12 months 18 months maybe um even though the tools have been around forever.
• 42:56
  SPEAKER_04: So now that totally changes the game because there are brands that are straight up creating content at scale in an automated way to just dominate search results.
• 43:05
  SPEAKER_01: Like if you can imagine it's being done it's possible it's being done by one of your competitors so um when you think about the barriers uh oh and I I did want to mention uh you do have this report the state of the report uh and that does talk about some barriers to entry and what why don't you talk a little bit about that report and how people can get it and uh and just what some of the topics are that would be of value to to the audience.
• 43:35
  SPEAKER_04: Yeah so um Marketing AI Institute um late last year uh partnered with Drift who is kind of one of the bigger probably one of the companies to watch in this space.
• 43:46
  SPEAKER_04: I think they just were valued at like unicorn status after a recent funding round.
• 43:51
  SPEAKER_04: But they make AI powered conversational marketing tools.
• 43:54
  SPEAKER_04: And we partnered with them to see if we could better understand this Gap, right?
• 44:00
  SPEAKER_04: Literally the questions you were asking is, okay, I've heard about this stuff.
• 44:04
  SPEAKER_04: Why aren't people using it more?
• 44:06
  SPEAKER_04: Because they should.
• 44:07
  SPEAKER_04: And so what are the gaps?
• 44:09
  SPEAKER_04: And we had a lot of hypotheses about it.
• 44:12
  SPEAKER_04: We thought, okay, maybe people just aren't there yet.
• 44:15
  SPEAKER_04: Maybe they're too afraid.
• 44:16
  SPEAKER_04: There's a lot of things out there that could scare you about AI.
• 44:19
  SPEAKER_04: So we decided to do a survey and then write a report about where we saw the state of marketing AI today.
• 44:27
  SPEAKER_04: So the report is you can find it at stateofmarketingai.com.
• 44:33
  SPEAKER_04: It's just free for download.
• 44:34
  SPEAKER_04: But long story short, we surveyed more than 400, 400 marketing leaders across a different number of functions, big enterprises, small companies.
• 44:43
  SPEAKER_04: The full breakdown of that audience is fully transparent in the report.
• 44:48
  SPEAKER_04: And we asked them a series of targeted questions about the their understanding and adoption of AI, like understanding how many were how many people were adopting it, what their barriers were to adoption.
• 45:02
  SPEAKER_04: And some really interesting threads came out of that.
• 45:04
  SPEAKER_04: So the biggest one is that it turns out when it comes to barriers to marketers wanting to adopt AI, fear is not really a huge factor among this audience.
• 45:18
  SPEAKER_04: So only about 15% of people said that fear was actually a barrier.
• 45:26
  SPEAKER_04: What was more interesting is about 70 plus percent said that education, a lack of education and training was the biggest barrier.
• 45:34
  SPEAKER_04: They knew it was important.
• 45:35
  SPEAKER_04: We asked other questions about that were meant to gauge are you, do you actually know AI is important to your business?
• 45:42
  SPEAKER_04: It's very clear marketers expected, even five years from now, many of their tasks to be intelligently automated or augmented with AI tools.
• 45:53
  SPEAKER_04: They didn't know how to get there.
• 45:55
  SPEAKER_04: They didn't know what tools they should prioritize, what use cases, how to go about actually using and piloting these tools.
• 46:02
  SPEAKER_04: And furthermore, we asked some questions around how companies provide training to their uh to their employees, and very, very few were doing anything like that.
• 46:13
  SPEAKER_04: So that was actually a really interesting finding.
• 46:16
  SPEAKER_04: Um, this this thing that basically marketers are mostly at this beginning stage of AI, and they're ready to move on.
• 46:23
  SPEAKER_04: They just don't know how to go from point A to point B.
• 46:27
  SPEAKER_02: But I think that's also, you know, you can't, I mean, you you, you know, we talked about this where you actually have now some really, really good defined tools that get them there as a no-code type of scenario and provide, you know, recommendation type of scenarios, right?
• 46:45
  SPEAKER_02: Instead of, I mean, if you're gonna, if you're gonna go out there and say, hey, listen, we want to um, you know, there's nothing, we've not seen anything other, then you will need a data scientist and somebody that actually can look at the data and actually put some Jupyter notebooks together and you know, and really just you know figure out what needs to happen.
• 47:01
  SPEAKER_02: But that's a long, long, long process.
• 47:04
  SPEAKER_02: And so I think, yeah, like to your point, there's definitely now tools and technologies, especially across speech, language, and vision, to give us the opportunity to look at not only content that we've created, but also post-production content that's actually you know in production and actually look to see how it's being used and then derive that information back.
• 47:25
  SPEAKER_02: So we can have a closed loop, right?
• 47:26
  SPEAKER_02: And that's I think the important thing is to see how important in your mind is having a human in the loop to validate or invalidate or you know, or some of those scenarios, like augmented AI with a human in the loop type of scenario.
• 47:39
  SPEAKER_04: I'll be honest, I don't think we're not gonna ever have that.
• 47:43
  SPEAKER_04: Um, I just think that these tools are going to advance to capabilities we probably can barely even dream of today.
• 47:49
  SPEAKER_04: But I do think, unless we're talking some very far future kind of sci-fi scenario, that a human will always be required to be in the loop and a human always should be in the loop.
• 48:01
  SPEAKER_04: And I think that that's going to be a critical skill of employment in the future, is understanding how to work hand in hand with machines in this way.
• 48:13
  SPEAKER_04: I think it's also going to be a key piece of strategy is mapping out where are your humans in the loop?
• 48:19
  SPEAKER_04: What do they do?
• 48:20
  SPEAKER_04: What are their responsibilities so that these systems you get the most out of them?
• 48:24
  SPEAKER_04: Again, it's like there's probably a very small handful of these systems where it's like you want a human in the loop so it doesn't run amok, but it's more having a human in the loop to get the most out of it and to understand there are things the human needs to be doing to get value out of these systems that the machine just can't do.
• 48:43
  SPEAKER_02: Well, I think of scenarios like subtitling, right?
• 48:45
  SPEAKER_02: Subtitling, if you're in an area, and you know, especially for compliance and you have to have a hundred percent compliance, there's no ASR engine on the planet today that will produce 100% accuracy, especially in a real-time scenario.
• 48:57
  SPEAKER_02: So from that perspective, you know, there's if it's user driven, you know, user-driven content, that's one thing, because then you know you can say, okay, well, you know, we have a platform that does subtitling against this.
• 49:07
  SPEAKER_02: There's a lot of those little guys out there.
• 49:09
  SPEAKER_02: But if you're in a really heavy-duty production environment for subtitling and for accessibility and things where you have to have 100%, you're gonna have to have a human in the loop every single time, right?
• 49:21
  SPEAKER_02: So, you know, but I think to your point, what we're talking about is the the fear, uncertainty, and doubt of that A is gonna take over my job is not isn't should be thrown out the window.
• 49:31
  SPEAKER_02: Yeah, the key here is that it's there to augment and create better efficiencies for people to do more with less, right?
• 49:39
  SPEAKER_02: Create more efficiencies, create more opportunities, have the ability to look at more content and the exceptions, not the rules, right?
• 49:48
  SPEAKER_02: So anyway.
• 49:50
  SPEAKER_01: That's great.
• 49:50
  SPEAKER_01: I I think this is very valuable.
• 49:52
  SPEAKER_01: Thank you, Mike, so much.
• 49:53
  SPEAKER_01: We're right about out of time.
• 49:54
  SPEAKER_01: Chris, thank you so much.
• 49:55
  SPEAKER_02: And yeah, hey, Mike, where can people find more information about you?
• 50:00
  SPEAKER_04: Oh, great, great question.
• 50:01
  SPEAKER_04: Um, you can find me really easily on LinkedIn.
• 50:03
  SPEAKER_04: That's a great way to reach out.
• 50:04
  SPEAKER_04: You can also email me, Mike at pr2020.com.
• 50:09
  SPEAKER_04: Happy to answer any questions or point you in the right direction.
• 50:13
  SPEAKER_04: Also, just go to marketingaiinstitute.com.
• 50:16
  SPEAKER_04: We have honestly over a thousand articles probably at this point on the subject.
• 50:21
  SPEAKER_04: So please uh avail yourself of those resources.
• 50:25
  SPEAKER_01: Thank you for that.
• 50:26
  SPEAKER_01: And uh and there should be some show notes as well where we can have those links and other resources.
• 50:31
  SPEAKER_01: Again, tremendous uh uh session.
• 50:33
  SPEAKER_01: Thank you so much.
• 50:34
  SPEAKER_01: Thanks for everybody, and uh we will talk to you next time.
• 50:41
  SPEAKER_00: Thanks for listening to the Early AI podcast with your host Seth Early and Chris Featherstone.
• 50:48
  SPEAKER_00: We hope you enjoyed the program and took away some ideas that you can start to implement in your organization right now.
• 50:54
  SPEAKER_00: If you're stuck and need help, check out www.early.com for information about early information science existing services.
• 51:02
  SPEAKER_00: And of course, check the show notes for links to any resources we mentioned today.
• 51:06
  SPEAKER_00: That's all for now.
• 51:07
  SPEAKER_00: See you next time.