59: Why Most Companies Are Using AI Completely Wrong (with Bill Takacs)

Bill Tacacs: I also think a really interesting thought experiment is I need to hire someone. Okay, what's that person going to do? Is that somebody who's doing a lot of administrative work? And if so, is there a way to put some type of AI in an agent or some automation leveraging those tools where you don't necessarily need to hire right now?

Polly Allen: I think there's a lot of organizations that are treating it, they just don't know how to act with this step change. So they're like, well, let's use AI to do existing processes faster or more efficiently.

Bill Tacacs: I have personal project I worked on. I'm still working on that. It was revolutionary in terms of like, oh boy, it's game change changing.

Polly Allen: Welcome to another episode of AI Product Leader. Every week we shine a light on what it really takes to join and thrive in the AI world. Real conversations with real AI builders, from founders to product leaders, AI experts, technologists, and everything in between to understand where the world of AI is right now and what that path looks like, whether you're building AI products or a career in AI. I'm your host, Paul Polly Allen. I'm the founder of AI Career Boost where I spend my time helping product leaders and executives navigate the world of AI and thrive, leading AI teams and product building. I um, spent my career in both the software development and product sides of the fence when it comes to developing products and I'm an ex Alexa AI Principal Product manager where I led the very first generative AI answers on Alexa back in 2020. I am super excited to be joined today by Bill, Bill Tacox. He's a veteran product executive who has spent more than two decades building and scaling category defining platforms across SaaS, security analytics, AI and enterprise infrast. He's currently the VP of Product at M M Train, which we'll be talking more about where he leads the product and engineering with board level accountability, scaling the business to 10 million plus ARR and cutting platform costs by 50%, doubling the velocity and launching ML driven risk intelligence products adopted by nearly a third of customers. So having AI actually adopted is something we'll definitely be touching on. Bill is a proven 0 to 1 operator and a turnaround leader with a track record that spans Salesforce, AOL, Instant Messenger, O'Reilly Media, HP and multiple venture backed companies. He's also a former U.S. army Ranger captain. So Bill blends disciplined execution with bold innovation, bringing clarity, accountability, operational rigor to complex organizations at scale. Bill, thanks so much for coming and sharing your experience with us today.

Bill Tacacs: Yeah, thank you and ah, wow, I appreciate that introduction.

Polly Allen: Wonderful. Well, that's a lot to unpack as far as like the journey you've been

Bill Tacacs: on in your career.

Polly Allen: We'd love to get to know you just a little bit better. How did you get, uh, into this space of technology in the first place, end up at so many vaunted organizations?

Bill Tacacs: When I left the service, I knew I wanted to work in tech. Uh, my family was living in California, my parents. So I came sort of before Silicon Valley was a thing as it is today, but still, you know, sort of foundational in terms of computing and what we know is tech today. And so I came out here and got some good advice from my dad, which I actually took, which was like, go learn to be a salesperson. So that's what I did first for a couple years, got some formal training

Polly Allen: there and then just start from a product perspective.

Bill Tacacs: It was so fortuitous if I uh, sing same thing to young people. You know, you really learn. Rubber meets the road in terms of being a salesperson. Having to carry, you know, have a plan, carry a quota. It was super foundational and very relevant to product. You're talking to customers all the time trying to figure out what their needs are and how your product fills that needs. And so unbeknownst to me at that time, this is all sort of looking in the rearview mirror. You know, you learn that it's a pretty core skill to have. In addition to being able to figure out how to help customers fit their needs through some personal contacts, I met some contacts at HP and that turned into a position there. And I started in their IT organization which was a super forward looking organization. So this is mid-90s. Yep, I have some gray hair, but they had already were already doing things like backhauling voice traffic on the Internet. Had uh, had already put Internet in all their offices, had started networked everything together. So they and were very lean and very cost conscious about it. So they were very thoughtful. Their plan. I came at a time where a lot of people say the old HP was there. So Lou Platt was the CEO trained by Bill and Dave. So that whole sort of construct about build things that last, make things that customers love and delight them, um, and learn that process. So through what I worked on in the IT organization, I got introduced to one of their product teams and that eventually turned into a position and that was sort of kind of the start of my career.

Polly Allen: Amazing. It's like an internal transition like a lot of people recommend if you want to get into product from Outside, a lot of people do kind of that transition inside an organization.

Bill Tacacs: That's right. It's not dissimilar to an engineer being an engineer for a while and then transitioning to a product person kind of doing. And a lot of what we were doing at that time was pretty advanced in terms of what an IT organization would be doing. So like m, we rolled email out to the whole company in like under like right around a year. And so that was 180,000 people got that to work back in the day when it was client server as an old term from the past.

Polly Allen: Yeah.

Bill Tacacs: So good stuff.

Polly Allen: It's interesting to think too that like you've been through kind of two of the largest changes in terms of like the big shifts in technology that are affecting our society. You know, I'd love to hear more about how that first shift in Silicon Valley, you know, you saw things changing and how that relates to what's happening today.

Bill Tacacs: I have a really good personal anecdote for that. So one of the positions I had before I joined HP is I worked for an Internet service provider and doing more of the client onboarding, some of the sales, a little bit of some of the networking things. The, uh, people were working with though, had been in and around working as engineers or technical people for quite a while in the Valley. And they both told me, hey, if you're going to be in tech, get ready for these cycles. Because they boom and they bust because they had lived through like chip cycles. I'll just say personal computing, you know, and they're like, you'll skyrocket and you can make a ton of money and then there'll be times where you're probably unemployed, looking for a job. So just buckle up. And so from a less about the technical standpoint, but just you as a person, that was just some foreshadowing of a lot I think, of folks, technical careers where you go work someplace, things go great and then, you know, if you work in a startup company and things don't hit, you've definitely lived through this before.

Polly Allen: Yeah.

Bill Tacacs: And so it was just, it was yet another really successful thing that are really fortuitous thing that it was just right place, right time and had some good, you know, I developed some good mentors from that.

Polly Allen: Amazing, amazing. So then you've been in product then, since then?

Bill Tacacs: Pretty much since, yeah, uh, since right around, you know, early double odds. Working in a product that touched big companies, was very well deployed and really learned. HP is definitely engineering heavy company. So, uh, learned really early on about you know, translating customer needs and requirements, working with engineers and having to get stuff on the street. And this was back in the day where we used to distribute our software on, uh, tape, you know, digital tape. Yeah, literally that. Yeah. I, you know, one of the innovate, quote unquote innovations we had is we actually put up an FTP site where you could download, you know, download things back in the day. But that was revolutionary back then and like, our competitors weren't doing it, but we were able to do it. And some of the things you don't think about from an innovation standpoint, like pricing and packaging, running analysis to find out that you're kind of nickel and diming folks to death and if you make these changes, you might sell more and then running experiments and tests to figure that out. And I had early experience in my career doing a lot of that and you know, again, I was part of a team and with some really good leaders that were, you know, had the autonomy and also the mindset to go do these things and do it in a very, in a way that I think applies a lot to AI today. So, you know, what is your hypothesis? What experiment? And of course this, I think this is just good product anyway. What are your hypothesis? What experiments can you run to validate your thinking? And then you make and then decide,

Polly Allen: and then decide based on data. That's so interesting though. Um, it sounds like you were exposed early to not just like, hey, this is frontier tech, but, but also just a culture of innovation. Like we're going to try and do stuff that people haven't already done before.

Bill Tacacs: My first, yeah, my first year to that point, my first year at HP, I think over 60% of the products that were on the street were invented, quote, unquote, invented the year before. So it was an innovation heavy culture.

Polly Allen: Yeah, absolutely. And so how essential do you think that is now? Is that more essential than ever?

Bill Tacacs: Oh, for sure. Yeah, for sure. And you know, I think we talked about this before, but I do now play this narrative here. You know, I think that there are definitely going to. If you're an existing company and you're not jumping into AI, I think it's questionable. Your survivability is going to be questionable because I think that there's just, there's going to be a small percentage. I'll just say 20%, he's our friend, Dr. Pareto, you know, 80, 20 rule there, that will survive because they'll make the transition and then there's just going to be a whole host of AI Native companies that have lower cost structures because they can automate things. It's more innovative. They know the technology because this is, it is game changing revolution. It's a revolutionary technology. It's not iterative step change. Yeah, it is a step change. You know, going from a, I'm trying to think of a good example of like networking itself. You just think about how revolutionary it was to go from a computer on your desktop, many people may not remember this, to a network computer. And then as you, you know, so that's sort of one big revolutionary step. The you know, a big step change. Getting it faster is not necessarily as revolutionary.

Polly Allen: Yeah, that's so true. And I think there's a lot of uh, organizations that are treating it, they just don't know how to act with this step change. So they're like, well let's use AI to do existing processes faster or more efficiently as like the most imaginative thing they can think to do or know how to operationalize.

Bill Tacacs: Yeah, I think you know, so I think you know it is one of the, I've spent uh, part of my career, uh, one of my in the IT organization hp. I was part of the team that was building workflow automation and so we like re engineered the petty cash process there. We got rid of cache windows and like put fraud controls in. We did. It was a major change to how things worked. Very fortuitous with AI. So yes, AI can actually automate and reduce a ton of repetitive tasks. But that I've made an analogy to folks. It's kind of like when AWS first appeared you were like oh, this is an easier way to host. Well not necessarily. That is sort of the wrong frame to think about about how to use it. It's how you put all of the components that are available to you in unique ways or innovative ways versus the traditional. I'm going to a server and have an Internet connection and run software on it. I'm now going to uh, build products with all these disparate components. And so with AI it's not. I think there's a good analogy there in that yes AI can go automate a bunch of tasks for you, but there's other innovations that AI can do that you're probably not even thinking about right now.

Polly Allen: Good way to frame it and actually a great opportunity to dive into more about what you're doing at M Train. And so what there's the kind of the bucket of hey, how do we work differently and how do we get people to adopt this internally but also in terms of like how do we put this into your product? Which would you rather tackle first?

Bill Tacacs: Yeah, well, so I'll start to just continue that brain. So, you know, I think we're like a lot of companies so, uh, you know, we. Let me back up. So in the summer of 24, we actively looked at building a. An agent, what we'll call an agent today, that was connected to a client's account that would do not only answer support questions, but sort of be a. An assistant with our product to that client. And we just opted not to do it at that time for a whole host of reasons. For a whole host of reasons. It was everything from cost to this may turn into a distraction and it's not clear to us that we're going to be able to have the security and safety that we'll need given what we do.

Polly Allen: And just so people know, what does M Train do? What's the two sentence?

Bill Tacacs: So we do. We are an um, E learning compliance training company. And so we do anti harassment, code of conduct, cyber security, those courses. We also have an analytics product where we can leverage the data from training to help you identify risks and errors. Organization. So around themes that we trained around the content we trained. So, you know, we can help you understand, you know, if you might have a claim that could appear in a team. But more importantly, what we. All of this is related to skills, social skills. And so we have an entire skilling framework that our content and our training is all related to. And so our analytics product maps into these skills and so we can tell you where you may be weak on a skill and that is something that might be leading to a risk area for your company or on your team.

Polly Allen: Absolutely. Yeah. Perfect. Excellent. So in terms of that space, compliance is such a riddled with regulations and things like that. How do you treat something where people are not used to 100% not being on the table for how systems are acting for when we're introducing.

Bill Tacacs: Yeah, so that's, you know, that's something that we're. We're still working through. And that's one of the, you know, one of the things that we were like M Chatbot, maybe not right now. Chat Asian, maybe not right now. What is required for us to understand that if we give a response to someone, it's not going to hallucinate. We also had, and I think a lot of companies heard too. Do we really need AI to solve the problem? Can we just solve it through basics? You know, I'll just say the basics of, you know, ML practices or just Connecting the dots from basic data, like tabulating something and data analytics or something.

Polly Allen: Yes, exactly right.

Bill Tacacs: And so, you know, where is it, where are we really going to apply AI to help people use our product more effectively and make it easier for them? Like a lot of companies right now, I think we are just, we are individually using AI, so it's kind of like using AI and let's observe and figure out it's not that super expensive to buy your teams, you know, one of the foundational AI companies tools and just see how far you get. So that's sort of one bucket. The other bucket is definitely seeing on the product life cycle and development, life cycle, software development, life cycle, AI acceleration. So let's experiment with these things and then I think we're at a point right now where we're starting to quantify their impact. So what's the impact on product velocity from all these things? Where are holes? Can we apply AI to solve some of the bottlenecks we might have or holes we might have? And so that's sort of that internal, at least in the product space, also in the company overall, now that we have some experience and people are more comfortable. You know, a lot of this is just folks using these things, like, oh, this is actually really great as opposed to like, I heard all these AI horror stories. It's going to hallucinate, it's going to do bad things. That's not the case. It's also not a magic bullet. You know, talk to any developer, draw

Polly Allen: that line right between like, there's the skeptics and then there's the, like, everything's amazing. And you're like, it's not. I know you tried it 15 times and, uh, looks great.

Bill Tacacs: That's right. Well, in content you develop, I like the developers because they're very quantified in what they do, you know, so it's like, yeah, this is definitely accelerating my work, but it's not doing my work for me. Now I do think the other thing that we have to keep an eye on in AI is that might have been true a couple months ago, but I'm seeing more and more success and I'm hearing more and more stories about engineering teams that are. All they're doing is running agents. And so, you know, the code agent is off. I have an agent writing or working on it, was responsible for this part of my code and I'm having it work on this thing. And there I have agents working on another part of my code and it's off doing this thing. And then I have a bot that's like doing all the CDCI effort. So I think that is very true. We are going to be in a world of agents that we basically own and manage and they do work on our behalf.

Polly Allen: Yeah, I think like a year ago that was still. There was a question, right, of like, is that science fiction and how true will that be and how quickly. But certainly in the software development space, I don't know if you've seen it on your product teams or things yet. I don't see it as much. Can you talk a little bit of that? Is it still really like software development is this constrained space? They've got a constrained language that they're working with. Right. In terms of like a vocabulary. It's. They've got more training data. Right. It's like a slightly more tractable problem than just do anything a. An employee could do or a product manager could do. Do you see as much automation and like that full like autonomous systems working out?

Bill Tacacs: Not all at to still inside? Yeah, we're just not uh, we're just not there yet in terms of, you know, definitely inside of our organization. Other peers I talk to, same sort of thing, partners that we might engage with, you know, same sort of thing. A handful of CTOs that I've been in contact with probably over the last quarter or so, they're kind of looking at it from the context of a. Somewhere in the neighborhood of a 30 to 60% increase in terms of productivity on their team. So. And I'm thinking about it as a product person in terms of what's our product velocity. So are we putting more things on the street is what's doing to quality? Because I do think we have had some instances where engineers have gone back through the code and the AI has helped refactor things that would have take like one. We would never have had time to do that. They were in the code. So they, you know, part of this was an experiment. Let me see what it will do. It's like, oh, ah, lo and behold, I had success. So I think there's again, we're still in a lot of experimentation phase. I. There was an article I think I saw this week where somebody like their engineers are not writing software anymore. That was sort of the narrative, like the engineers don't write software, the agents are writing software. And I'm like, okay, is that klarna? The pay as you go company was a big meme last time this year where they had laid off like some massive amount of their support people. And six months later they're like bringing them back because it's, you know, it's just not there. Plus, if you're going to interact with a human, human's really gonna want to, want to talk to humans. Especially if I'm, you know, trying to negotiate a bill, I guess. But, but no, I think that, that it's going to come and I think it's going to come faster than folks think it's going to be there. There's also a lot of pieces being put in place to be able to facilitate that there with the open claw, you know, sort of whole open claw attention that's happening around it right now. You know, there are people who are working on the ability for whatever you want to call it, a robot, an agent, a replicant, to have its own wallet and be able to transact. You start to think about crypto technology

Polly Allen: or online shopping changing completely overnight.

Bill Tacacs: Digital currency microtransactions in that context become viable in a way they aren't today. And so you can certainly see agent to agent communication and exchange of value in that context happening. So. And how fast will that come? A lot of people trying to solve these problems. And so. And of course the, the AI is really accelerating all of it.

Polly Allen: Absolutely, absolutely. It's, it's interesting to see how quickly stuff is moving, particularly on the tech and software engineering side. And then it's like there's this separate world of folks who are like, I tried ChatGPT six months ago, it wasn't that good.

Bill Tacacs: Yeah, well, you know, this, you know, this all goes back to like, do you know how to, do you know how to use it? And you know, yes, it is a better search, but that's not, you know, it's not a pattern. I have. And I think you, I'm sure you've noticed this too. The more that you think about it as an expert in, you potentially are interacting with that you have unlimited questions with that can help strengthen your ideas, your plans, your thinking, you know, in that sort of context. Not my ideas, but stuff that, you know, I certainly experienced it, but, you know, those types of things, plus the automation capability is really great. So the other thing I am witnessing and I think this is going to be, it'll be, uh, I have no idea whether this is going to really pan out, but the collapsing of roles and the product team, product and engineering team, for sure I can do design, I can crank out designs and prototypes where I would probably have to do some kind of wireframing handed off to our designer. They'll put Some polish on it might be the reason we would do that is to go socialize something with a client so it kind of looks, you know, they can really thoroughly understand the experience. That was something potentially if we were going to do it was an effort, we'd have to scope it, you know, we'd have plan it. Now it can kind of happen on the fly and that's a collapse. First time in a long time I have have written code. Well, the robots have written code. So I basically prompt things, able to commit, get peer reviewed and so that by no stretch of the imagination a developer in that context. However, you know, if you need a label change, I can like product people can probably do that. If you see like a button broken, you could probably fix it in the future. And so that, you know, that's going to be really design. I mean, you know, it's like here's my wireframe. Can you make it look like this? Like that was not possible without an exceptional amount of effort. Now it now is. And of course it's true the other way. And of course some things you know, also to think about is like how do you leverage AI in the context of going to talk to your customers or your clients to get.

Polly Allen: Yeah.

Bill Tacacs: You know, doing research for you. How do you instrument your product to use it in that way? So there's just a ton of use cases that AI can probably be applied to that will collapse roles, change roles, those types of things. I should mention a couple of things since we started so kind of thinking about things in a kind of two buckets. So one is like how do we take AI and make ourselves more productive, more efficient? How can we do more with less on the AI side also not spend ourselves into oblivion.

Polly Allen: Another reason to think about OpenCL people,

Bill Tacacs: we all hear it's like I'm spending a thousand, you know, a thousand dollars a day on tokens for my open claw, you know, replicant. I mean it, you know those things are happening anymore. Yeah, yeah. And so you know, compute costs and tokens. It's all real. And so you have to think about where you're going to automate and where you're going to get the most bang for the buck. A lot of that just starts. One of the things we've started with is what are your daily tasks? So what are you doing on a. All day, every day or once a week and are there steps to that task? Very, very definable steps. So you know, I open a browser, I log into this thing, I get this data, I put it over here, kind of basics of workflow. So there was a foundational book written back in the day by a guy named Michael Hammer called Re Engineering the Corporation. And so, you know, he's like, everything has a process. There's a process everywhere. Start by just following the CAL path. So follow the path, just write it down, just interview people, write it down, you'll figure out the process, then you can automate. I think those principles of automation workflow have not changed. They're definitely apply. And so that's a good way to start thinking about AI. Uh, and a way to start experimenting

Polly Allen: on your product teams like efficiency or hey, where can we exactly make some adjustments or just magnify what we have kind of thing to.

Bill Tacacs: That's right. We do a lot of digital paperwork, writing tickets, validating tickets. So you know, intake, you know, feature intake. So how can we apply AI to make all of that, all of those chores of moving things around easier? And that is a basic, a very basic case. You could also argue, uh, don't necessarily need AI to do a lot of that. It's a good way to start working with it.

Polly Allen: That's it. And just thinking about what to automate because I found that too even like as a solopreneur, there's all this advice about like, oh, and build all these systems. You're like, if the process is going to change in a week, like don't. That maybe is not the thing to invest in. And that's maybe that discernment that we still, that a lot of companies are lacking.

Bill Tacacs: So I also think a really interesting thought experiment is, oh, I need to hire someone. Okay, what's that person going to do? Is that somebody who's doing a lot of administrative work? And if so, is that, is there a way to put some type of AI in an agent or some, you know, some automation leveraging those tools where you don't necessarily need to hire right now. And I think, you know, Open Claw is getting a lot of attention because it's making, making those types of automations easy for people, you know, securely. Well, that we haven't figured out yet.

Polly Allen: Right, exactly. Um, so there's that first bucket that's all about like improving efficiency on your product team, on your team in general.

Bill Tacacs: The second bucket, the second bucket is, you know, really how do we build product with AI? Uh, so what is our answer? And our CEO founder Janine Yancey did a really smart thing and just said, just started to go top down and say we're going to rebuild these things with AI. And we're going to put AI in them. Those things are our authoring tool. So how we author content because we have a proprietary system to do that and we're going to make improvements and changes into the user experience for how we deliver content. And so we've gone from zero to one quite quickly with a very small, with a couple of people, like literally a handful of people. And so uh, I do think it is one of those things that, you know, if you have an innovation mindset, you're going to have to do these types of things, things to break the log jam of like AI. I'm working with AI. It's great to make my job faster, easier. It looks like I'm literate and I can write better now. Like those kind of things, pedestrian use cases to like how are we going to really make a change for our clients and ourselves. And so our authoring tool, eventually we will eventually enable it, enable our clients to use it where they can create their own content or remix our own content. Other things we're looking at is we have a pretty significant library, uh, of workplace related videos. We are video heavy on our training and so what are our opportunities there to put AI in, to remix, um, and to work which we do today. But it's all, you know, it all has to be, you know, a human has to do it. And so we get requests like, oh, this is a factory scene. Can you put it in an office? Can you make this an animated video six months ago? Those things are super expensive. Not really clear that could happen. You know, by the time this hits in a quarter or two, it definitely is going to be a uh, capability. We'll have to figure out the economics for it because there may be, that may change sort of how we turn it on for folks. And then our vision, our vision really is to the point where you need to be compliant from a harassment standpoint, anti harassment. So just go do everything for me. I could just talk to an agent that will do it and so it'll set that. It'll figure out who needs to train on what content to be compliant. It'll do, it'll set up all the training campaigns and it'll run it. It'll do things like hey, here's the naughty list. I call it the naughty list of people who are behind and managers have, you know, managers, leaders have their own

Polly Allen: them to do their compliance training.

Bill Tacacs: Email summary, um, reports like all of the stuff you go in and click buttons to do today. Can we just take that off the table for people Absolutely right.

Polly Allen: I love this use case because it's a great use case because things like, hey, like follow these checklists, follow these processes. It's a nice constrained way to use AI in a way that is less, that has less of the determinism and hallucination problems. Right. Where you're just like, we can actually make sure that it's really following, it's getting you to follow these rules. But I especially love that it's also, it's not like anyone that loves following up with emails with the people who didn't do their compliance.

Bill Tacacs: That's right. That's right. Issue organizations, you know, the automation of continuing to remind you, I'll, you know, I'll use a negative nagging, you know, nagging you to get it done. You know, it just takes it, it makes something that has to happen that is a chore and automates that chore and does it in a way that's probably more friendly, more interesting and just, you know, if you're an HR person and have to do this now you have an agent robot doing it for you, you can do higher order things for your order.

Polly Allen: Absolutely. Then how do we augment, uh, people and have them working at a higher level opposed to replacing them?

Bill Tacacs: For sure. Yeah, for sure.

Polly Allen: Amazing. Uh, well, it's incredible to see, you know, leading a company like M Train through this both in terms of like getting those products out to market quickly and changing internally how they're working. How do you recommend other folks who maybe feel behind on AI want to get started? How do you recommend that they get started in the AI space if they feel like, hey, I need to get this my hands around this?

Bill Tacacs: Yeah, so just start, just identify something. A good thing is, like we talked about earlier, identify some basic workflow you're doing right now and just go to start. And if you have a team, get your team together and say, what are the top three pain point chore things we do all the time and see if you can automate them. And there's a number of ways to go do this. I'll just say, just start. I'll also highly recommend anybody who's in the product space, join AI Career Boost and take one of your things. It is an accelerator in terms of being able to leverage experience of somebody who's been found foundational in this. So your experience at Amazon, foundational in terms of sort of the thinking also, you know, having to live through hallucinations and the impact of all of that. And as you've talked about in some of the Classes. It is one of those things that's definitely worth paying for some guidance and help. And of course, you know, an organization like AI Career Boost, you can get mentor, ongoing mentorship. So that's super helpful as well. And it's very approachable and also digestible. So it's not a bunch of like, you know, here's a bunch of technical stuff that you're going to have to learn. It's pretty practical and again, just get started. Trying to automate things I think is a good way to begin.

Polly Allen: Yeah, getting, getting hands on while getting learning at the same time is really the goal. Thanks for the shout out. Absolutely.

Bill Tacacs: Oh, you bet. I also have to say I was on a different podcast webinar, I want to say like three weeks ago with Google and Meta, uh, pm. I think one was, you know, director level product managers there and one of them, unfortunately, I'd love to attribute the quote but you know, the narrative was basically no one has this figured out right now. No one really knows exactly what the patterns are. No one really knows what they're doing. So you're either going to start or you're going to get left behind. It really is one of those, it is one of those things and you're going to have to give yourself and your organization some time and space to experiment. So just like you would with your product team, if you're a senior leader, you need to be thinking about that approach as well is like, what are the top three things if it was automated, would really make an impact and let's go try to do them. And if it, you know, again, it's very much a startup mindset. Fail, fail fast. You know, that's very true. It may not work. And then, you know, if you're going to put it into product, as I mentioned, we had looked at doing an agent chatbot years ago, but between cost and safety, we just weren't there. So. But I wouldn't, wouldn't have even known those words at that time. It's just, you know, as we progress through that process, they emerged in terms of like m, is this thing going to be reliable? Like what happens if it, you know, we haven't. Part of the things we do is you have to have a capability to ask an anonymous HR question in some geographies. So we have this whole ask an expert. We're like, well, what happens if we turn the expert on? And it's like, you know, has a very draconian, you know, my boss is harassing me. I'm going to make A very pejorative thing up here. My boss is harassing me and the chat bot comes back and says, we'll go do all these, you know, talk manager, talk to HR team. You come back, no, none of that's working. And it's like, we'll hit them with a bat. Yeah. Like, I mean, it could hallucinate like that. You know, that. So, you know, we were like, oh. You know, we're like, oh, well, that. But we can't have that happen. And so those are. You're not. What I'm trying to say is you're not going to understand for your situation how to even begin thinking about this unless you start to try to do some of it, you know, carve out your time.

Polly Allen: Time, absolutely.

Bill Tacacs: Uh, and.

Polly Allen: And I think that's. That's where that innovation mindset we were talking about is so important is a lot of people who don't come from very innovative companies are uncomfortable taking goals or taking on projects that they don't know ahead of time will succeed and want to do all the research ahead of time.

Bill Tacacs: It's very. Even with our own organization where, you know, you got set into a motion that, uh, motion is working, it's successful, it's progressing, and a change shows up like this, you as a leader really need to intervene and just be aware that you're gonna introduce some chaos into the, you know, some chaos and change. And I don't mean that in a negative frame. What I mean is it's gonna be a change and it's gonna feel uncomfortable and chaotic to people. But I think if you're an incumbent company and you're not doing it, there is probably somebody. Whatever you do, there's probably somebody working on an AI first version of it that has lower cost structure because they have less people, they can go way faster. And you have. Have to. Yeah.

Polly Allen: Need for leaders who can lead people through this is so urgent right now. That's right.

Bill Tacacs: And the basic leadership skills applies. This is independent of technology. So, you know, the right frame, the right framing, you know, the right sort of thinking about outcomes. If you come to Polly's class, you'll learn about watermelon goals and why those are important. But all of that aside, it is very much sort of leading. In the military, they talk often about leading from the front. So being in the front of the organization and really being able to see what's happening and understanding what. What's going on with the situation. And so it's very relevant here in terms of you're going to have to take that first step. You're probably going to have to pull people because you're going to be at the front and it's not. If you're uncomfortable, it's probably right.

Polly Allen: I love that. Oh, that's a perfect, that's a perfect quote. I love it.

Bill Tacacs: Feels like a wool coat and it's scratchy and itchy. You're probably on the right track.

Polly Allen: Oh, that's great. That's uh. Oh my gosh, I love that. That's fantastic advice, Bill. Thanks so much. So we have time for one last section and that is plugs.

Bill Tacacs: Yeah, uh, for sure. I'll plug my company for sure. So if you need harassment training, code of conduct, cyber security, we have a pretty deep catalog. We have those offerings. But more importantly, we can help you not only with the training, but leverage our risk intelligence product to really uplevel your skills in working in your social skills, in your organization. And it's when you start to think about it, it's like what do you mean? It's all skills, skill based and when you think about the skills that are here. So it's managing power, team dynamics in group out groups. If you don't know what those are, they're. And you're a leader, you should, you should understand those things and we can help you not only understand them, train on them, but help you understand where you may have areas of opportunity to improve in all of it. And it's a, um, more unique in that way and that we do. Our training's great. It's mostly video based so it's not horrible. People love watching our videos and they're a little hyperbolic in some instances because they're trying to reinforce a point. But they're, you know, they're of available out there. So yeah, please come. We will put my email address in all of this so you can come contact me there. I can introduce you to the right folks and I'm trained if you're interested.

Polly Allen: Wonderful, excellent. Got all of your clients and education needs. Fantastic.

Bill Tacacs: I'll also shout out to you and what you're working on here. AI career boost if you're in the product or develop realm for that matter. But seriously investigate it. It's worth having gone through it myself. It's worth the time and effort. It definitely, definitely was an accelerator for me.

Polly Allen: Oh, uh, fantastic. I remember the first time you let me know you'd done our three prompt protocol to build your first prototype. I think it was m. Your first.

Bill Tacacs: I sure did. I had personal project. I worked on. I'm still working on that. And it's like, yeah, it was revolutionary in terms of like, oh, boy, this is, this is game. It's game changing. And so my whole little narrative there about AI first, AI native companies versus incumbent, that really solidified it for me because I was able to do things that would have taken me, uh, years, quite frankly.

Polly Allen: And you're like, wait, if this is in everybody's hands, how do we need to operate?

Bill Tacacs: Yeah, it's like, whoa. And, you know, we hear a lot about sort of the single founder unicorn, you know, a, uh, two to three Decacorn, two to three person Decacorn companies. Those things are all gonna. They're coming, they're coming. And with Open Call, you're starting to see a little bit of like, maybe

Polly Allen: no people, company

Bill Tacacs: know, you know what, there could be some, you know, that's, you know, will the AI develop its own sort of organization? Who knows?

Polly Allen: So how would they organize that?

Bill Tacacs: Oh, my goodness.

Polly Allen: Yeah. Excellent. But yes, do get a hold of Bill. We will put his email in the show notes and so you can go and check out M Train from my end. I'd love to plug our, um, masterclass upcoming. Masterclass Path to AI Product Leadership. That may be where Bill first ran into us in our classes, but we go through, hey, what are the emergency capabilities we need to be aware of as product leaders, Director plus product leaders? What are the skills we need and what are the things we need to get hands on with to be able to lead from the front, like Bill was just describing? So that'll be coming up in the next couple weeks. We'll put that in the link in the show notes there as well. Thanks again so much, Bill. Sharing your experience and actually from someone who is operating on this and actually wrestling with these problems today and where you're at, it's so great to hear about that, like, on the ground experience and how you're managing innovation within your company, especially with the constraints of complex compliance and play. Thank you. Yeah, thanks so much. This is wonderful. That's our episode of AI Product Leader this week. We'll see you all in a couple weeks. Bye for now.