It’s the hottest topic since the “world wide web” disrupted everything for businesses. Artificial Intelligence (AI) has been around for quite some time already, quietly humming in the background, having a nearly imperceptible impact on most individuals and organizations – until in late 2022, OpenAI released ChatGPT to the public, and our collective minds were blown.
As the whole world turns its attention to the new frontier of innovation, plenty of speculation exists about whether AI is a friend or foe to businesses and their employees. Will AI be a threat to artists, authors, musicians, analysts, developers, and all other creators and knowledge workers? Or does AI hold the promise of maximizing our human potential and expanding our capacity for positive innovation?
While many questions remain to be answered, one thing is certain: AI is here to stay, and like all technology disruptions, the competitive advantage will go to the early adopters who embrace rapid change as a catalyst for growth.
In this recorded webinar featuring a panel of experienced B2B and B2C marketing leaders, we explore the practical and powerful advantages of AI for entrepreneurial growth organizations.
Key Takeaways
- Marketers need to adapt to changing and emerging technologies such as AI.
- AI is empowering small businesses to be more agile and adapt to market needs faster than ever before.
- AI doesn’t always produce 100% accurate results – so use it as a starting point.
- The best way to get more comfortable with AI technologies is to jump in and “get your hands dirty”.
Resources Mentioned
- The power of the Authentic Mindshare™ and CMO peer community
- AI Prompt Engineering Isn’t the Future
- AI Readiness: What Business Leaders Need to Know
- Pictory AI
Full Webinar Transcription
Introduction and AI Overview
Jennifer Zick: Welcome, welcome to another Authentic Growth™ webinar. We’re going to give all of our attendees just a moment to roll in. I see the attendance meter ticking upward as people are making their way into the zoom room. So we’ll wait just a moment to make sure people have a chance to log in and join us. But I just want to say welcome. If you’re here, you’ve made it, hopefully to the right place. We’re here to talk about embracing AI and what that means for all of us in marketing and business growth. So we’re excited for a great conversation today and so glad that you could be here with us. We’ll kick things officially in just a moment. So in the meanwhile, grab a if you’re at your desk, grab a notebook, grab a pen.
I’m sure there are a lot of notes you’re going to take away today, and we’d of course love to hear all of your feedback as well. So. All right, well, we are 1 minute into the show and we want to conserve all of that space. So we’re going to go ahead and get started. And I want to start by welcoming everyone again. I am your host, Jennifer Zick. I am the founder and CEO of Authentic®. We are a fractional CMO team that works with growing businesses all across the US and sometimes beyond, to help them Overcome Random Acts of Marketing® and confidently take the next right step toward healthy growth. So most of the businesses that we’re working with are between five and 100 million in annual revenue.
There are all kinds of industries and business models, but the one thing that a lot of our clients have in common is that they are really working towards strategic growth and scale. And I think that this question about AI is on a lot of people’s minds right now about how can AI help to accelerate growth and where are the benefits and where are the risks, and how do we need to think about using AI within our growing businesses. So we’re really excited to talk about that with all of you today. And we know that there’s a really wide variety of attendees on today’s webinar. We’ve got business owners, we have investors, lots of founder and entrepreneur types. We have marketing leaders and their team members, agencies.
So we know you all have a different lens through which you’re going to have an interest and perspective on today’s topic. For the most part, we’re going to address it as if we’re speaking to that entrepreneur founder of a growth business. But we hope every one of you is able to take a little nugget out of today’s conversation that gives you something actionable or something interesting to further explore. And we would love to continue being a resource for you as you deepen your exploration of AI, because we are on the same journey as you are. So before we kick off the topic and dig deep into the content, I want to give all of my esteemed panelists an opportunity to introduce themselves. I’m here today with three other CMO colleagues from Authentic. And Ruth, I’d love to just start with you.
Say hello to our audience and tell us a little bit about yourself.
Ruth Glaser: Great, thanks, Jen. Hi, everyone. I’m Ruth Glaser and I’ve been a fractional CMO with Authentic for about three years. And I’ve worked with a variety of clients, both B2C, B2B, from technology to e-commerce. And what’s really exciting to me is to see all the different ways that AI can be applied to any business and is being applied in virtually any business across all industries. And every day I find myself saying, wow, that is so cool. That is so cool. That’s like my mantra lately with regard to AI. I’m really excited to be a part of this conversation today.
Jennifer Zick: Thank you, Ruth. John, over to you.
John Ryan: All right. Hi, my name is John Ryan and I’ve been in B2B high tech marketing and sales for over 30 years. I’ve lived through the beginning of the PC era into distributed computing, into the Internet, and to the day that we’re seeing today, which is an extension, I believe, of digital transformation and really excited to hear what my colleagues had to say as well as to try to be of service here today with what I’m seeing.
Jennifer Zick: Thank you, John. Hey, Peter.
Peter Zaballos: Hi there. Hey, thanks to everybody for sharing the afternoon with us today. And I’m Peter Zaballos. I’ve been with Authentic for two years and I spent my career in much the same trajectory that John did. Technology companies, my whole career in marketing roles. And today with AI, it’s like the most exciting and most overwhelming transformation I think we’re going to see from a technology perspective in our lifetimes. And I know everybody said that about the Internet and they said that about the iPhone, but what we’re seeing unfold with AI today is going to pretty fundamentally reshape how everybody gets their work done and it’s going to be fun today to talk about what we’re seeing in marketing.
Benefits, Challenges, and Risks of AI
Jennifer Zick: Absolutely. Well, thank you all so much for sharing your wisdom and sharing your time, and thank you to all of our audience members for carving out your time to be with us here today. I want to invite you to participate on the audience side to be part of today’s conversation. You’ll see in the zoom controls that there is a Q and A feature. The Q and a feature is where you’re going to want to ask any questions that you would have for any of our panelists, either individually or collectively. And we’ll work to create some time toward the end of today’s panel conversation to address some questions that you may have. The chat feature is just more like commentary, visible to everybody, but if you have a specific question you’d like us to tee up to the panel, please use the Q and A.
And this session is being recorded, so don’t worry if you have a colleague that you want that isn’t here today or you want to revisit it, you’re going to get the recap as well as a few other resources in our follow up email after the event today. So. Well, let’s dig right in. And I want to start by finding my first question, which is for Peter. All right, Peter, let’s set the stage, because at Authentic we have this thing called Mindshare. It’s how all of our team, all of our cmos gather together on a regular basis through structured meetings, but also ongoing collaboration through our Slack channels, to share our learnings, share our wisdom, and share all the things that we’re exploring in marketing solutions. And AI has been probably the biggest channel in our slack environment over the past several months.
Like when chat GPT was publicly released, it was like our collective minds have been blown and we’ve been digging in and exploring, but AI has actually existed for quite some time. It’s kind of been humming along in the background. There’s been a lot of different modalities being tested, but it’s with really the release of chat GPT that’s really gotten the world to wake up and look at it. So AI is a super interesting topic, but where are we at in the overall maturity of this category of AI? And what are some of the benefits, challenges and risks right now when it comes to AI overall?
Peter Zaballos: Well, that is an awesome, broad topic to sort of set the stage for what we’re gonna do today. And you’re absolutely right about AI having been around for a long time. It has. It’s been around for over a decade and longer. But what’s changed is a confluence of three things coming together at once. One is the quality of the mathematical modeling and specifically large language modeling. You’ll hear the term LLM thrown around a lot, and that’s just a way of describing the kind of language model that’s being used to interpret what you’re asking the engine to tell for you. So the quality of the modeling has gotten a lot better. The increasing power of hardware has sort of met that at a really opportune juncture. Specifically, GPU’s graphical processing units are really good at interpreting large language models and AI.
So those GPU’s have gotten super powerful and reasonably affordable. And then the last thing is the size of the training content that these models have been able to access. Because an AI, an LLM, is only as good as the content you train it on. And if you think about what it takes to be able to answer a simple question, it takes an awful lot of training to answer a question that could be phrased in 25 different ways, that could be using terminology specific to a region or a domain. So this last topic of training these models is the one you probably read about the most, because a lot of the training got done before the people that had all the content understood the value of that.
And right now, today, there’s a big controversy with Reddit because a lot of training was done by people accessing Reddit APIs and taking all that content and running through the models. And Reddit stood up and said, hey, I don’t know if we gave you permission for that and maybe we should charge you. So what’s happening now is the mathematical models have gotten really good, the hardware’s gotten really good, and they’ve accessed enough content to train these so that they can answer simple and complex questions with a tremendous amount of accuracy and do it in a way that’s very conversational and interprets things like tone. So that part is super powerful and super important. But to say, where are we right now? We’re still very early in the process of this.
It’s almost a good analogy when Steve Jobs introduced the iPhone. He probably anticipated that there was going to be an app store. At the time, he said there wasn’t, and eventually there was. But at the time that phone was released, it didn’t exist. App Store didn’t exist. I think we’re going to see that AI and large language models are going to be a core component of almost everything we do. And that is going to change a lot over the next four to five years. The most visible one that’s out there now is, as Jen mentioned, Chat GPT. It’s made by a company called OpenAI. Their version 3.5 was pretty good, but when they released version four, the training they had done was so much more extensive that it became incredibly good.
And the introduction of 4.0 is what really blew this market up. And while it’s true that LLMs and AI in general are going to fundamentally reshape how we do our jobs, that it’s not perfect by any means, and we’ll show you a little bit later about where it does a really good job and where it still has a lot of work to do. In the meantime, our jobs are all secure. Like, I think what you’re going to find throughout this talk today is that AI can be an incredibly productive tool for you, but it doesn’t replace you. It’s just going to change how you do your work.
Jennifer Zick: Thank you for taking the anxiety right out of the airwaves right off the bat, Peter, because I am guessing some people are joining this call wondering what’s going to become of all of the makers and marketers and creatives, and we’re going to deep dive into that as a specific question later. But John and Ruth, would you like to take on at all on that initial question and what Peter shared in terms of benefits, challenges, risks, where are we at in the lifecycle of maturity?
Ruth Glaser: Well, I think of it for those of us who were around when the Internet became ubiquitous for the first time, it was like when you could create a website in HTML, like everybody could do it, and you’d have a rainbow of colors and flashing fonts and things like that. And it feels very much like we’re in sort of that infancy stage of regular people or all people applying it in different ways. So we’re all starting to learn this is a really cool tool. How are we going to use it? I’m also starting to see tools that we’re already using incorporating AI in more forward facing ways. So, for example, HubSpot recently gave me an alert that they have a beta version of an AI content creation tool.
So we’re going to start to see it being applied more visibly in our everyday tools and work that we’re doing.
Jennifer Zick: Right. I think most of us have only truly experienced it platform wise as the chat robot. Right. So now it’s going a lot deeper. John, did you have something to add?
John Ryan: Yeah, I want to add on to what Peter had to say. I don’t mean to put the anxiety back in the conversation, but he was talking about these large language models. In those models are what’s called parameters. Those are variables in the system. And to give you an idea of what’s going on is you’ve already been using AI. It’s just artificial narrow intelligence, not generative intelligence. So when you use Netflix or Amazon and they make a suggestion, that’s Ani AGI is what we’re seeing now. And so what to give you an idea of what’s happened in our lifetime, which is, I think, quite wonderful. I think this is an inflection point for us. We just need to manage it properly. But when you look at deep blue, it had 8800 parameters.
It had under 9000 parameters, and it was specifically put together to play chess. If you look at GPT-3 that’s 175 billion parameters. If you look at GPT, I’m sorry, if you look at Google Bard, that’s 540 billion parameters. If you look at version four, GPT, that’s estimated to be 1 trillion parameters, and then version five, which is coming out next year, is going to have 17 trillion parameters. So I wanted to give you an idea of the kind of power we’re going to have, but I would like you to position it this way. It’s decision support for you. You cannot put your judgment on the sideline. It is to help you make the right decisions, not make your decision.
AI Applications in Marketing
Jennifer Zick: That is a really great landing spot on that one. And we’re going to move on to the next question. And so many of the other parts of this conversation are going to build upon the foundation you all just set with those perspectives. So, Ruth, I’m coming to you with this one. I’m reading a lot, and I’m hearing a lot from our CMOs. You’re all out there fueling a lot of learning for me, and I appreciate that. Tell us about AI prompts, what this skill set is. That’s called prompt engineering. Why is it going to become, and is becoming so important for marketers?
Ruth Glaser: Sure. Well, so to get quality results from AI, you have to give it a very specific set of directions. And prompt engineering just means using the right words, phrases, structures and so on to guide AI to get those good results. I have also found that it requires additional prompts to refine the output. So it’s not as though you just input what you need, and the result is perfect and ready for prime time. So marketers need to get really good at creating prompts. And fortunately, it’s amazing how quickly this has gone. Even in the last couple of months, you’ve just seen an explosion in examples, courses, learning materials on how to get good at engineering prompts, if you will. And then, of course, you think you’re on top of things with engineering prompts.
And then you read, like I did a Harvard Business review article that says prompt engineering is on its way out because it’s going to become obsolete. The AI systems are actually evolving and getting better at understanding natural language, so you don’t have to be quite as specific. And then the new AI language models, like with the GPT four that was mentioned, they’re already improving, right? So they’ve got templated prompts. So AI itself is on the verge of making prompts less important. And what this article in particular suggested that we need to get really good at is identifying problems and formulating them in such a way that we get the objectives accomplished through AI. So really learning to identify those problems and the core problem is going to be an important part of leveraging AI in the future.
Jennifer Zick: That’s smart. That reminds me of even in using essential reporting skills or working from your CRM and driving a report, you need to know the question that you’re trying to answer in order to leverage the technology, and AI is really going to accelerate that and bring new insights to the surface faster. But Peter, John, what would you add? Peter, you’re muted.
Peter Zaballos: No, I think getting to the answer quicker is exactly what this helps you do. And I like John’s comment earlier about this becomes your ally, it becomes your trusted resource, but it helps you get specific and insightful about things and saves you a tremendous amount of time in the process. And maybe that’s something that we could keep coming back to. What this helps you do is save time. You can rely on it to go and do work and investigate things for you in a matter of seconds. That would take you hours to go do yourself.
Jennifer Zick: Right. We were just talking in my Vistage group yesterday about optimizing human performance and teaching the skill of delegate, automate and eliminate. Right. And I think AI is going to be the answer categorically for a lot of those ways of solving so that our impact is elevated.
John Ryan: Yeah, I love what Peter and what Ruth said. What Ruth said there you can take away and say, I can work off of this. I think that, you know, I’ve read two books on how to ask better questions, and now I want that time back. I think that what we want to get really good at is asking the best question we can, whether we’re talking to AI or talking to a colleague, to get the best answer we can. And those who ask better questions get better answers.
Jennifer Zick: Right? So how long if AI is already getting smart enough to take away prompt engineering requirements, how long before it’s telling us what questions we ought to be asking?
Ruth Glaser: It does that right now. You can ask it, how should I ask you for this? And it will tell you to a certain degree.
Future of Marketing with AI
Jennifer Zick: That’s fascinating. That’s fascinating. All right, well, Peter, taking it to another angle here, we started talking about some of the. Oh, wait, I didn’t want to skip one. John, I’m coming back to you because, Peter, you’re holding on to a little bit of a demo for this next one. We’re not quite ready for it. Yeah, let’s look ahead to five or ten years. We already said, or we at least heard a perspective saying our jobs are not going away. Marketing is not going to become an extinct category. Right. The creators are going to live on. But how do the roles change? What is marketing and creative and the makers, how does that change?
John Ryan: All right, well, great question. So I start this off with a joke, which is to say that, and I’m sure you’ve heard this joke, is two people are in the woods and they see a bear, and the bear is coming at him. And one of them takes his backpack off and starts putting on his sneakers. And he says, what are you trying to do? You’re not going to outrun that bear. And he goes, I just have to outrun you. The way I would look at this is you got to get on it. You got to, you’re going to be using it directly, as Peter and Ruth have articulated, and you’re also going to have it embedded in the Martech and the sales tech that you’re using. So it’s going to be pervasive.
And the reason why is because, again, it’s an extension of digital transformation. So when these tools are coming out, there’ll be an expectation of, well, what AI do you have in the system? Which leads to, oh, I’m on Wall Street. I need to explain why I built this technology and how it’s going to grow our business. You’ve got to be able to answer that question. It’s all connected. There will be large investments in AI going forward, and you’ve just gotta keep that in mind. But, hey, here’s the deal. If you’re a progressive marketer, you’re going to keep doing what you’ve always done. You’re going to work alongside your technology and do better.
Jennifer Zick: Thoughts from my other panelists on this?
Peter Zaballos: Well, I would just add, in a lot of respects, nothing has changed in marketing today in digital marketing, especially what you were doing a year ago, is no longer effective today, no longer relevant. That marketing today is a matter of constantly evaluating new technologies, new approaches, using data to tell you that the thing you were using before is not effective and you have to do something else. So this is exactly in that theme. But using these tools, you’re going to be able to obsolete yourself much quicker and become much more capable much more quickly. So in a lot of respects, I think this is just a, what happened was this showed up so fast, a lot of us were like, wow, now what?
But the effective marketers are also the ones going, okay, well, it’s a new thing, I better figure it out, I better learn how to do it. And I think that’s where we are. And it’s just a little shocking to have this literally come out of nowhere. Raising a whole bunch of other questions too. Like Microsoft is replacing their search engine with chat GPT. At what point are we no longer going to be using traditional search engines and we’re going to be using this? I think that directionally we’re going to be using large language models instead of traditional search engines. It’s just going to take some time, but it’s coming and we just have to get used to it and learn it.
Jennifer Zick: Another fascinating trend that was raised by our speaker at my Vistage meeting yesterday. His background is employment law attorney, but he advises on a number of talent related issues. And the current trend is toward a skills based workforce in the sense that traditional titles and rank and level of experience are mattering less and less because the world is changing faster and faster. So now there are more organizations moving to flatter titling structures and elevating where the skills fit the needs of the business. And this seems like a real game changer in that regard.
Ruth Glaser: Yeah, I think so. I think the other thing that it does is it really empowers, especially smaller businesses and smaller teams to start doing things that previously would have required a lot of expenditure or a lot of headcount. So I think in the next 2510 years, we can’t even imagine how this is going to transform our workplace. I mean, even the things that you see as examples now, people are so smart and amazing at getting the tool to do incredible things. It’s going to be really exciting, I think. And the other thing that is really fantastic about it is it frees us up to do other things. I mean, if this can take off our plates, some of the repetitive tasks that can and should be automated, that frees us up to work on higher level projects. So it’s all very exciting.
John Ryan: Yeah. The thing I like about Ruth and Peter and the thing I would ask any of you to think about is, what is my adaptability quotient? I understand what my IQ is. I understand my EQ is what’s my adaptability quotient? Can I change quickly and pivot and go in the direction that will benefit me, my company, those around me, and move with the technology, as Peter pointed out. So I would ask you to have a good conversation with you about that because I know that Ruth and Peter have high adaptability quotients, as you’ve heard here, and that can really help your career and the way you run your business.
Capabilities of AI in Marketing
Jennifer Zick: Absolutely. And it really throws me back to when I was first entering my marketing career in the late nineties, early thousands, and I was working for a small web development agency, and that was where I was starting to see. And I really fell in love with the way that the emergence of the universal use of the World Wide Web was letting small businesses compete at scale and disrupt global enterprises. Right. And that same opportunity is there with this new wave of technology, just like it was with the last wave of digital transformation. Those who are nimble enough to make the change faster and being smart about it are really going to gain a competitive edge. And the challenge for larger organizations is being able to turn a titanic.
So the innovation, the heart of innovation, entrepreneurial teams have a lot of advantages there, I think.
All right, Peter, I jumped ahead, but now I’m on track. Back on track. All right, so we started by talking about some of the overall capabilities of AI. But I want to get specific. I want our listening audience and our viewing audience to see some specifics. There are already hundreds, I don’t know, thousands. I don’t even know how many tools there are that are leveraging AI. And there are so many ways to apply AI personally and professionally. I’d love to know what you’ve been observing and learning in terms of the most practical applications, the game changers that we can adopt now. And what are you learning about these new tools as they emerge?
Peter Zaballos: Well, the main thing I’m learning is to just use them. And, you know, going back to what was talked about earlier, you have to start thinking about, you know, what prompts are you using and start experimenting with it. Right now, this whole topic area is so noisy that you can quickly get overwhelmed because, you know, there’s a new AI tool coming out every minute. And then all the existing tools are quickly implementing AI based enhancements. What I thought I’d do is just use chat GPT four with some real examples that I recently used with one of my clients. One was we wanted to hire a digital marketing content writer. So the way I would normally go do that, and I’ve done it a million times, you all have to.
You go to, Indeed, you go to LinkedIn and you get a couple of examples of job descriptions that are kind of like the one you’re interested in. And then you spend an hour merging it all together and getting it to be coherent and then getting your own voice in it and having it speak specifically to the culture of the company. So I’m going to share my screen and show you this position I just filled what I was able to do. You can see this now. I just went to it and said, write a job description for digital marketing content writer in an early stage technology startup who sells cloud based software, and you can see what it’s doing in real time.
And the part that just blew me away about this was that it’s adding a whole lot of content that a human would have wanted to add. It understands early stage technology software companies, so it’s adding terminology relevant to that. I no longer require college degrees and instead have equivalent work experience. So even though it’s saying bachelor degree in marketing, which is journalism or related field, this brings up another topic about how to use AI. This is generating an 85% complete document for me. And now I need to go back and say, well, I don’t require college degrees. Now what I did was take all of this.
What you just saw saved me an hour, and then I could spend ten minutes crafting it and tuning it for how I would go and hand it off to HR and say, can we go push this rec open? And that’s literally what happened, you know, and then I thought another area that we’ve been using this is to take a long white paper. So I’m just, and you can see I’ve already done this. So I’m saying, can you generate, actually, I’m going to go up and say, because this is another area, can you generate a hundred word summary of the following article with a conversational voice and from an executive’s perspective? And then this is a long article that we wrote, and I do the same thing. And, and it’s come back with this long white paper about the modular housing market.
And it’s created, it’s not 100 words, it’s a little over 100 words. If I go back up into this history, you can see where I went up to it earlier and said, can you generate a 25 word summary of the same article? And this is what it came back with. That is not 25 words. And it can’t say, I can’t do that for you. It just tried and did its best. And what I’m sensing is that it looked at that fairly complicated copy and it used its language modeling to understand context, and I was only able to compress it down to about 100 words. And a human would need to do the 25 words because that’s an awful lot of interpretation. And I did this job description for somebody we hired a few months ago.
And one thing you’re seeing is that as more people are using these large language models and more daylight is getting shed on how they are constructed, where’s the data coming from? What type of data can I rely on or not? They’re starting to enhance this to give you a little bit more visibility into the limitations. I was the CMO at two companies before I joined Authentic, and the last one was this company called Cumulo in Seattle, and it was a file storage company. I’m just going to regenerate this prompt and say, who are the primary competitors of Cumulo? It’s a private company, venture backed. So the prompt that comes back now, it’s saying, oh, guess what? My knowledge cut off is September 2021. So already it’s telling you, okay, this is 18 month old data.
But then they come back and it knows what Cumulo does, and then it comes back and lists all their competitors. And these are their competitors, and they have been chat. GPT has framed the essence of these competitors’ value propositions really succinctly. But I was the CMO at a public company called Cumulo, I’m sorry, SPS Commerce, who’s in the cloud based supply chain business. And I gave them the same prompt and same disclaimer. The date is September 2021. And now I’m looking at this, and I’m going, number five and number six are actually competitors. Everyone else is not. In some cases, they’re partners. In some cases, they’re not even relevant. So you can’t, especially when it comes to, like, quantitative data, it’s really limited.
And I know someone who was planning a vacation to a coastal city in Texas and asked it to make hotel recommendations, and it gave five recommendations, and two of them, the hotels, did not even exist. It made them up. So this goes back to, in all these examples, this can save you a ton of time, but it does not replace your judgment, does not replace your experience, but it’s a really efficient tool that lets you get more done more quickly, but you can’t just nail the results in without scrutinizing them and ensuring that they actually are what you need.
AI Not Producing Accurate Results
Jennifer Zick: That’s right. And Ruth, you shared with me a couple of personal case studies that you’ve run into on where robot generated AI content was not accurate or helpful or well positioned. Can you share a couple examples of that?
Ruth Glaser: Well, AI can’t count. I’ve requested a certain amount of words on topics, and it sometimes produces things that are wildly different for exactly the reasons Peter explained. It will make things up as well. Like Peter said, there was an example I read recently where an attorney filed a brief that cited previous court cases that were non existent. So, yeah, you don’t trust it. You have to double check, particularly if it cites anything specific. You can ask for sources so that you can check them out. But, yeah, it’s still at that stage where it’s not 100% accurate all of the time.
Developing AI Skills in Marketing Teams
Jennifer Zick: So right back to you again, Ruth. How can employers begin proactively to develop AI skills within their teams? How can they think about AI capabilities in their ongoing hiring and further building out their business and up leveling?
Ruth Glaser: Yeah, you know, I asked AI this question right before the webinar, and it came up with some pretty good suggestions. So AI, particularly in marketing, is really useful. And I think a good way to level up your team skillset is to just dive in and start using it. I think you just play around with the tool, encouraging your team to do the same. So many of the AI tools available right now are free or have trial offers. I played with one earlier this week called Pictory AI that you can use to create videos. And I was so excited about how quick and easy it was to create a decent video that I shared my draft with the team. I mean, it wasn’t ready for use publicly, but it was just exciting at how quickly it could be created.
And then that sparked the team to start trying it out, and then they shared their example. So I think that just having that environment of curiosity and learning and experimentation and then making sure you’ve got ways for people to share their learnings can be really helpful. I know that we do that in our mindshare that Jen mentioned earlier. And if you’re a leader, maybe you make it a specific time that you set aside on a regular basis for the team to share their learnings. The other thing that can be really important is cross departmental collaboration. So I guarantee you that if you’re in marketing, experimenting with AI, you can be assured that your IT counterparts are doing the same. And if you’ve got data people on your team, they’re doing it as well.
So having those conversations cross departmentally about what they’re learning and the applications that they’re doing can really be helpful. And then there’s been such an explosion of information and examples that just Googling, training resources or examples, you’re going to get a million hits that you can start to dive into. But I think once you catch and your team catches the AI bug that it just starts to snowball.
Jennifer Zick: Absolutely. And I can imagine. We know that when it comes to creating value in a business, speaking now to the business owners and founders and investors on this call, we know that a lot of what is looked at in terms of acquiring a business or valuing a business is how productive is the business in terms of where are they using technology and automations, and is it productive business that’s driving more to the bottom line by way of use of those kinds of technologies? I think there’s going to be a huge opportunity for businesses to drive more to the bottom line by creating more productivity with their existing talent and not needing to hire an extra person to fulfill whatever the next administrative task or research task or whatever that might be, which is going to create more transferable value. That’s going to be fascinating.
Anyone else on this topic of how businesses can think about developing skill and why that’s important?
John Ryan: No, I can’t.
Peter Zaballos: I just think it.
John Ryan: Yeah, go ahead. Sorry, Peter.
Peter Zaballos: No, I think this is going to start putting pressure on organizations to invest in skills development for their teams, like everybody should. On every marketing team, you should be taking free classes on how do LLMs work and start picking a tool that’s relevant to your role as a manager. You should be encouraging everybody. I don’t care what tool you pick, just pick the one relevant for you and start getting good at it. Let’s talk about it, share it with the group. But this is going to take professional development up a level of importance, I think.
Excitement and Concerns Surrounding AI
Jennifer Zick: Absolutely, John, as we move toward this last question for our organized panel conversation, and we’re going to leave some time to engage the audience because I see that there’s some good questions coming in. So audience, take a moment, make sure you get your question in the q and a so we can have some time to address it for you. John, kick us off as a marketing leader and you started the conversation by sharing some of those massive milestones in marketing evolution that you’ve lived through and that most of us on this call have lived through. But when you look at AI, what do you find most exciting about it today and in the future? And on the flip side of that coin, what do you find particularly concerning or cautionary?
John Ryan: Well, I think something my daughter said to me last week, which really stuck with me, Washington, we can’t make the technology so good. We create bad people. And so, you know, we have to keep this in mind. We need ethics around this. We need to understand the power of it. We’ve already talked about how you need to maintain your judgment that you’re going to put on top of it. We’ve talked about that. So I think we need to be careful here, like we do with everything that has, you know, wonderful, great power, but in the wrong hands and the wrong bias written into the code can cause problems. So we have to pay attention to that. We still have to be in charge and pull the plug if we have to.
So the other part of this, though, what I would say to everyone here is understand where you stand. And where you stand right now, I believe, is, as Peter said at the beginning, an inflection point in history. This has been compared to the Gutenberg press, which before the Gutenberg press, the people that controlled information were monks and people who lived in castles. So we’re in a place now. We’re at the democratization of decision support. But here are the things that I would also share with you, because AI is a lot about math. IDC projects that we’ll spend over $300 billion on AI by 2026. Mackenzie is already showing, already saying that we’ll have a two to $4 trillion added to the economy annually because of AI. You’ve got Microsoft committing $13 billion to AI. Google, the story I saw was 30 billion.
Amazon, I think was 22. No, Facebook was 22 billion. I think Amazon was 10 billion. The big four accounting firms investing billions, Accenture 3 billion. What you’re seeing here is a tsunami and you want to catch the wave and you want to really understand what’s going on. The reason these companies go toward this, by the way, if you haven’t seen this in the past, and digital transformation is this is how they grow their stock price. They catch a big wave like this and then they build capability around it and they try to improve the companies that they work with. And so that’s what’s happening here.
We all need to be thinking about, as Peter and Jennifer and Ruth have said, we all need to be thinking about how we’re going to understand this and how we’re going to use it to our benefit and our company’s benefit. And it’s exciting. I think it’s a great time to be alive. I just think it’s very exciting and I’m looking forward to the conversation.
Jennifer Zick: Thanks, John. Peter and Ruth, what would you add?
Peter Zaballos: Buckle your seatbelts, take a grammar mean because. Yeah, I mean, this is a structural change in how computing supports the jobs we do and I the lives we live. It’s as fundamental as the invention of the PC or the iPhone and because it’s software, it’s going to be everywhere.
Ruth Glaser: Exactly. I love what John said earlier about adaptability quotient. I think that’s really important to embrace this change and not look at it as a threat, but such an incredible tool that we’re at the beginning of right now. My college age kids wonder how I got through college without the Internet. Their kids will wonder how they got through college without or with just the beginnings of AI. So, yeah, it’s coming and it’s exciting.
John Ryan: Can I say one thing there? I just want to say this quickly, Ruth, you’re spot on. We’ve been talking a lot in the last ten years about digital natives, ladies and gentlemen. We’re going to be talking about AI natives for the next 20 years. They’ll have different expectations.
Jennifer Zick: I will say the first time I ever heard the term chat GPT was through my daughter and her college friends who were working on coding and they were so excited about what it could do to accelerate their designing and innovation. It’s just amazing. We’ve got some really awesome questions from the audience and I know some of our panelists have even started to address them, but I’m going to surface a couple of them that have been chatter among ourselves in our mind share, and one of those are, is should we be thinking about developing policies, company policies, around the use of AI? I would love. This was a part of our mind share conversation yesterday. So Ruth, you led that conversation. What are we seeing? Because I know our clients are asking this question. We’re asking it with them.
Ruth Glaser: Well, I think we’re in the Wild west right now of AI because there isn’t any or much government regulation around it. I think there will be in the future as we start to see how nefarious some of the applications might be. I do know that in our discussion yesterday, there was talk of different companies developing policies so that they’re transparent in what has been created with the assistance of AIH. When there isn’t transparency about that it can lead to some controversy. There was an example given of a commercial that was created that didn’t disclose the AI assistance that was used to create it and that has resulted in some backlash against the company.
So I think it’s wise to think about developing policies for both how you want to use it internally and then how you disclose it when you do use it in a public facing way.
Jennifer Zick: Another example that was shared in our mindshare conversation was a brand that has started to watermark any images in their content that were AI assisted or AI generated. So I do think transparency is a key word because without regulation we’re kind of setting our own rules as we go. And I know the US is further behind on regulating than some other countries are around AI already. It kind of takes me back again on the way back to Napster and the original. You know, the Internet arrived and we discovered that we could stream whatever music we wanted easily. And then all of a sudden we realized maybe that’s not the ethical way to go about getting our music right. And there had to be some regulation on that. So I think there will be those evolutions and it’ll be interesting to watch that.
Peter Zaballos: Enhance that a bit. To say it’s a combination of Napster and Netscape. For those people who remember Netscape and Napster because Napster was taking somebody else’s content and distributing it. And the AI equivalent is somebody training on someone else’s content like Reddit and not getting permission and not paying for it. But it’s like Netscape in that when I was at another company, back when Netscape was around, I met with the CIO of a big defense contractor and he said he just got off the phone with Netscape and he said we have a policy that we don’t allow any web servers inside our firewall without the CTO’s approval. Netscape sales rep said, actually you got 756 of them running right now that you don’t know about because they can just get downloaded. Let’s talk about a site license.
And I think we’ve got both of that happening. AI is just, it’s like the tide, it’s coming in and there are some safeguards we need to be thinking about. But you’re not going to be able to legislate. Don’t use it if it’s going to be more effective for people. So I think it’s a combination of how do you embrace it and how do you ensure that you’re not exposing legal liability to yourself, like Ruth was saying, but boy, this is just going to go like wildfire.
Jennifer Zick: Well, and this leads into another question that at least two of our audience members posted, which is, what do we know about protecting intellectual property when it comes to use of these tools? Or I even think back to the live demo you ran us through, Peter, and you prompted with a couple different questions to generate a job description or an article summary. And I’ve played around with it enough to know that you could use the same prompt and get very different. Like, it’s not going to give you the exact same content each time, which in a way protects against, you know, multiple companies getting out there and writing the exact same blog and posting it. But on the other hand, you know, how do we create truly unique ip of our own and also protect our own ip while we’re embracing this technology?
Peter Zaballos: It’s going to be hard. I don’t think there’s a clean answer to that right now. I think it’s clean to be able to say if you’re creating music or you’re creating video or images, are you using an LLM that has a license for the training content? That’s really clear. I don’t know the legal implications of that query. I just made that prompt and that content I gave it. What happens to that content? And I don’t know the answer to that. I haven’t seen a whole lot written about that. You see a lot written about the large corpuses of content that they access and the legal implications there. But what happens to people that use it in the content they submit? I don’t know.
Jennifer Zick: Right. One of our attendees asked a question about, let’s say you’re working inside of your corporate email and you’re responding to a client’s question, and maybe you’re in sales and you get the same kind of questions regularly. Can you use AI to potentially borrow that content through email and phone call conversations to start to address on your website faqs? How might those things work together as you think about accelerating Q and a for your own business?
John Ryan: Well, if I understand you right, I think people have been doing that for a while. You should be learning from what your customers are asking you and imagining that other customers, like them in your ideal client profile, are asking the same question and write content around that.
Jennifer Zick: Yeah. Yeah. As a small use case, we had been using otter AI and just recently shifted to Fireflies AI because it integrates better with our HubSpot. But that tool follows me into all of my sales meetings. It follows us into candidate conversations, and it not only captures the audio with their permission, they’re alerted to it, but it transcribes it and it summarizes it in a really interesting way where if we chose to do so, we could aggregate a lot of those summary type details and learn from some of the common things that could be used, and that there are technologies that are centered on helping use those kinds of things for sales training and for website content. So I hope that answers the question that the audience member posed. And we’ve got time for maybe a couple more.
I’m just going to go ahead and dive in on this one. Any recommendations on how you can continuously validate the mix between AI and human input? As a college student back in the day, we were trained not to utilize others’ words. That’s plagiarism. So how do you recommend we continue to motivate our human selves to stay authentic? Nice word. While using this new technology.
Ruth Glaser: I think a healthy dose of skepticism goes a long way. So when you get a result just reading through it, for example, I had a result that said studies show that. And so then I asked AI to cite the source for that and if they can, and then I can check the source. The other thing with regard to plagiarism, when were first playing around with chat GPT, we ran it through a plagiarism checker and there isn’t overlap. And you can design prompts requesting that the content, for example, be original or create use this certain type of tone or these techniques to create an original piece. So there are some things that you can do on the front end to prevent that.
But I do think it’s always going to require that human skepticism and overview, like looking through things and really stress testing it, making sure that it’s valid and accurate.
Jennifer Zick: And I can tell you as the I would oh, go ahead, Peter.
Peter Zaballos: And I would build on that by saying, like when I talk to college students, I tell them one of the cruel twists of higher education is that when you’re in college, you get failed or kicked out of school for copying. And when you’re in a job, you get danged on performance for not copying. Like when you launch a project, the first thing I will tell people is somebody’s done this before, so go figure out who’s done it before and then let’s learn from them. Copy what was good there and then build on it. But don’t spend a ton of time reinventing the wheel. So in college, yeah, you can’t plagiarize, but in the work. You have to. That’s just modeling best practices. But what Ruth was saying is really true.
You have to apply a skeptical eye to what do you get back and does that make sense and does that look fair and what disclosures are merited by what you got back? But, yeah, you kind of have to. Best practices are just another way of copying.
John Ryan: I imagine that’s a great way of positioning that. Peter, I agree with that. How many times have we heard that don’t reinvent the wheel? I think there’s another site called perplexity AI where it will cite where it’s getting the information, which makes it a little easier, which you can try that, too. I think what you’re going to see is the use of AI against AIh to keep it in line. And you’ll start to see systems and applications written to take a look at what’s being done and to take the proper action because of what it’s seeing. And in the end, I think that there’ll just be a market for that. We don’t like this kind of behavior, or this doesn’t look good. We need something to detect that and to be able to take action.
Jennifer Zick: That’s right.
Ruth Glaser: You know, to build on that. John, I think it’s intel, where I’ve been hearing commercials recently, how they’ve identified technology to identify deepfake videos. So you’re exactly right that the checks and balances are in process right now.
Jennifer Zick: Yes. And I was going to add that as the mother of a college student, the one who told me about chat GPT and was excited to use it, she’s also been sharing with me how their professors are using AI to combat AI. So it’s out there. I think it’d be a tough time to be a professor and be grading papers, that’s for sure. But, well, we’re getting toward the top of our hour, and I think we’ve answered pretty much all of the fundamental questions that the audience has. So let me just do a last round of room and see if any of our panelists have kind of a closing mark, a last nugget of wisdom or curiosity or question you’d like to leave with our audience to take away. Peter, is there anything you want to leave them with?
Peter Zaballos: Before LLM showed up, I used to say this, there was never a better time to be in marketing because it’s all digital, it’s all instrumented, it’s all about data, and it’s just so great because you know what works and what doesn’t, and you can. Iterate your performance and now it’s like that, but just on steroids. It’s never been a better time to be in marketing. With all of this AI flooding into us, there’s going to be a whole new world of tools and effectiveness that we’re going to be jumping on.
Jennifer Zick: That’s exciting, John.
John Ryan: Well, I think we live interesting times. And how lucky are we? I actually, to be perfectly candid, I’ve been waiting for this day all my life. I think that our ability to use technology to help us make better decisions, look, we got a lot of problems we’re not solving as human beings on our own. We could use the help. So we need this kind of firepower to address some of the things that we’re dealing with. And the thing I would say as a marketer, be careful. Around three or 04:00 in the afternoon, you’re really tired. You rely on AI too much.
Jennifer Zick: Get chocolate and caffeine instead, and then get a fresh reset.
John Ryan: That’s right. Don’t let it get by you. When you get a little tired, you’re like, hey, I’m going to go to AI and see what it has to say.
Jennifer Zick: It’s fantastic. Ruth, what about you?
Ruth Glaser: Yeah, I would just encourage people to jump in and to start using it. And I’d also like to note that there are a lot of non business applications, locations too, where you can use it. Peter had mentioned trip planning, for example. And while you might get some non-existent hotels, you can also get a really great itinerary for a place you’re trying to visit. So lots of different ways, exciting ways to use AI. And I just can’t wait for what’s coming.
Jennifer Zick: That’s awesome. Well, I want to thank each of you again for sharing your time and your wisdom and your passion for innovation with our attendees. And I want to thank our attendees. And I want to invite you to stay connected with us individually. I know all of us on the panel today would be happy to link in with you to keep a dialogue going. We invite you to continue to follow Authentic as we stay on our learning curve in order to help guide our clients on overcoming random acts of marketing and continuing to grow in a healthy direction. And we would welcome any feedback or any questions that you have that might inspire a future webinar that you would love to hear about or some of our content planning we’re going to continue to ask you. We might go ask the bots too.
We’ll ask AI, what should we put in our next webinar? But we really value your feedback even more than that. So, on behalf of all of us at Authentic, thank you so much for joining us today, and we hope you’ll come back again soon. Take care.
Interested in learning more about the power of AI and how to prepare your business? Check out our comprehensive guide for tips and tricks.
