So as with most things in life, moderation or someplace in the middle is where you need to find yourself. The real the real through line, again, is that I think, and what you just said, that companies that skip the AI literacy don't want to be too biblical here, but they're building their house on sand. Because if everybody doesn't understand fundamentally what it is, why we're using it, and how we should use it, that's all part of the literacy, then it's not going to work.

I feel like that opening line for every one of them is this changes everything. But I think the the takeaway there at a higher level for us is that utilization of these tools, they're going to get better, they're going to get faster, they're going to get more dangerous over time. And those tools are going to ⁓ like ⁓ be subject to government regulation. And ⁓ You know, I mean ⁓ that that that's that's not that's that's policy and and you know, we don't need to get into that. But it it's also not unique, right? ⁓ over the last ⁓ couple of months, ⁓ one of the f tools that we were both using for a while, and I still use it some and I pay for ⁓ Manaus, which was bought by Meta, ⁓ which was founded originally by ⁓ A group of Chinese researchers and ⁓ and when they set their company up, they set it up in Singapore. So there's a whole thing there about like, you know, who owns the IP and and ⁓ but right now the Chinese government has said that even though that that purchase has occurred, you need to unwind it. Meta is in the middle of unwinding a purchase of billions of dollars, ⁓ where they've already integrated staff and teams and trying to figure out a way to To unwind this purchase ⁓ of a product that they've already baked into their product. ⁓ And we're probably going to see more examples, right? Whether it's telling a company that you can't sell or distribute, or that you can't be purchased, or that you can't roll this model out. ⁓ And this is where, you know, we started our conversation today talking a little bit about, you know, well, you know, some people are saying we should save money by running our models locally. Maybe. Or maybe it's not about saving money. It's about saving as a risk mitigation point of having you know the the the term here is inference. You're paying for inference. That's the token generation. Having local inference is is a is something that might be a fallback where you have to have ⁓ you know a model that's compatible with whatever ⁓ you know we talk about the frontier models, the ones that are the most cutting edge that whether it's anthropic or open AI or ⁓ things that like Alibaba and are are rolling out like the Quen models and and stuff like that, th those We ⁓ or or deep seek. ⁓ we need to be ⁓ aware of the fact that like this is something that we need to think about from a risk mitigation standpoint, right? Not ⁓ get caught up in

I'll give you one more story and then I'll kind of tie that together. So open AI, same situation, just a little bit different. So Anthropic seems to be dealing with the executive function. OpenAI seems to be dealing with ⁓ the I guess the judicial and the judicial executive function from a lot of different states. I think there was 42 different states that basically called open AI to the principal's office because they want to know hey It is It is very important that we understand how you're advertising to us, not just to us, but specifically to, you know, groups that are a little bit more vulnerable, maybe not able to advocate for themselves. So like kids and scene very, very similar to what we saw with Facebook. And there's a new movie coming out about that too. Looks kind of cheesy. the Facebook that is. There's a new movie coming out about Facebook in the fallout from that. ⁓

Seniors. Yeah.

And they want they want records on how we're handling all the data, especially the health data and how the engagement process works. So hey, here's three more things that we can do. How do they make it ⁓ sticky? There was even something in there about ⁓ syncophancy. So is this thing a syncophant? You know, how do you how do you get out of the chat bot what you need, not what you want? to hear, if that makes sense. So these are all this is all just kind of in the initial phases. But this is this is in preparation, in my opinion. This is one way that unfortunately this is being politicized too, is the way I

I s I see that too. Right.

Yeah, and there's without going into all of that, both of these folks want to IPO. That means they're public companies in a few months, a year. Now the scrutiny is ramping up before they get there. And both of them have a little bit different of a problem right now, but I think the problem impacts both of them. Whatever comes out is precedent that is now set for Everyone in the industry. And I do agree with what you said though. The through line really for all executives is hey, these models might change on a dime through legislation. And if there are changes, you don't have control of that. And so now this is a this is a risk that you do need to model in. Okay, what if I So I'm using Fable Five. What if I use Fable Five for this new workflow? And I need it because it does so much better at cybersecurity. That's what we're using it for. We went quick and we changed our entire workflow to use this model, or we spun up a workflow to use this model. And now, a week later, it's gone. And we had already rolled this out, touted it to our investors, let everybody know we're on the cutting edge, yada yada yada. Now you're going back and going, Well, Mayaculpa didn't have any idea this would happen. So I think that's a new paradigm. That's a new, that's just a new way that we're going to have to start looking at these models.

Yeah, the the this risk mitigation piece and and ⁓ the fact that ⁓ that right now there isn't necessarily for very specific things like this, we're from a security standpoint with like what what people believed that ⁓ this first this next model from Anthropic could do and can do. ⁓ there's not a a a fungible ⁓ alternative necessarily. ⁓ so how do you plan product roadmaps around you know, putting all your eggs in one basket, so to speak. and I think the thing that you were also saying about like these companies that were ⁓ they're being looked at not just for like the data and the privacy and the the these folks who are protected groups or ⁓ groups that are at risk. ⁓ but this idea of like how how responsible is the is the system around people, right? Is it is it going to ⁓ promote ⁓ poor behavior? Is it gonna pr promote ⁓ you know, activity or behaviors that we maybe don't want it to? And I think that's where, you know, this has been a fight with this is I maybe part of the the the next ⁓ Facebook movie that's gonna come out, right? Is this idea of how are we ⁓ influencing the general public by being able to control the messaging, how the message is delivered, how quickly it's delivered, who it's delivered to. what the voice is that they people hear. And I think the other like through line coming into the summer ⁓ is just general public opinion. ⁓ I've had ⁓ actually in the last just two weeks several conversations with younger folks. And ⁓ you know we we saw in the news at ⁓ there were a couple of commencement addresses where ⁓ was it Sam Altman or someone else, you know, like got booed. Yeah.

Even talking even talking to my kids, I mean, there's this big debate, it sounds like in the high schools over, you know, music and art specifically. ⁓ they do not like AI in music and art. I think they're okay with it augmenting. But philosophically or I think philosophically is the best way of saying it. They just they are at odds with AI, you know, taking over what are considered traditionally human endeavors. They're they love it for making funny pictures. And there's another really alarming trend, and this was in the Wall Street Journal today. There's another really alarming trend about nudifying photos, which is scary having kids myself. You have kids, yeah, where, you know, It's clearly not real, but it's another form of, you know, cyberbullying. And I think that's what this whole open AI thing is about, is how do we get ahead of this?

Yeah. And being able like those ⁓ you know it's one of those things that now that a bell has been rung, you you know, you can't unring the bell. No. and ⁓ you know, these ⁓ open models, ⁓ like you bring that up, like the idea of these that you know some of the tools, ⁓ some of the frontier models are trying to be better about you know, controls and guardrails and And some are trying to say, ⁓ no, you know, there's a freedom in being able to do whatever you want. And ⁓ you know, that that that you know it's not the responsibility of the of the model. ⁓ you know, you and I are both working with some some some self-hosted models. And as I look at the community that's ⁓ optimizing the the different models, there are there is a whole subgenre that is designed is built around removing the guardrails. ⁓ from the the the published models like ⁓ as a as a mathematical function. so that's one of those areas where I think the the through line there for us is not just about risk, but about perception. As we bring these tools into our business and into our ⁓ systems, ⁓ I think saying that you're not going to use them, that's not realistic. ⁓ Saying you're using them in a in a way that is responsible and trying to at least, you know, at a minimum have policy around it i is is gonna have to be part of the thing that we have to to bake into our organizations.

Yeah. There's going to need to be somebody who and probably the CIO or the CFO, I mean it's gonna take everybody, but somebody is gonna need to be responsible for understanding what is going on in the industry around AI and the impact it could have on their business, just like they would any other supply chain issue.

Yeah. The issue around AI is not whether you use you should use it, how much you should use it, how much you should tell people to use it, or how much it should cost. I think what we're telling both of our cli ⁓ you know, our clients, both of us are telling folks, use it, use it responsibly, ⁓ become fluent in what it's capable of doing, ⁓ become fluent in the terminology so that you can have kind of you know good strong ⁓ you know conversations across the table with each other without misinterpreting what ⁓ you know the words mean and and you know where costs lie or where danger lies or whatever ⁓ where risk lies. ⁓ But we're also ⁓ I think seeing the fact that we're we're in a place where Education. I I mean I'm I'm talking to several folks about it right now, about ⁓ having that language, understanding just a little bit more. You don't need to get into the super like nerdy matrix multiplication, you know, quantization of of of tools. I do because well that's that's that that's that's my space. But it it's ⁓ it's super interesting to see. ⁓ but it's more about The fact that these are tools that we can use to help solve and analyze problems, ⁓ whether it's supply chain problems that we can sh shine a light on processes by having really smart LLMs look at the supply chain, or whether it's you know financial forecasting and ⁓ you know, you know, ⁓ simulation type work. There's so much value. It's baked in there. Yes, we have to play this game now ⁓ as as executives, talking about how public perception is about these products, getting users to adopt it. And like I said, you know, I'm talking to younger folks who are there's a pushback right now. And so how do we get those folks over the hump of Whether you want to have a personal opinion about these products, great, but they are gonna have their place in the in the workplace. ⁓ and I don't think you're gonna get a a ⁓ a buy in the workplace by saying, Well, you know, I fundamentally disagree with with these these tools. Well, you're you're you know, fundamentally disagree with unemployment. Yes. I

Well, I I agree with you. So we we started by saying, hey, we've got this two pounds of AI problem. So executives are demanding AI without knowing what they're really asking for. We also talked about it being a hammer in search of a nail and some ways that you could get around that. From there we talked about token maxing, ⁓ Amazon burning AI budgets without a North Star. And how the opposite, so coupon cutting or going too cheap, that doesn't work either. So as with most things in life, moderation or someplace in the middle is where you need to find yourself. The real the real through line again is that I think, and what you just said, that companies that skip the AI literacy. Don't want to be too biblical here, but they're building their house on sand. Because if everybody doesn't understand fundamentally what it is, why we're using it, and how we should use it, that's all part of the literacy, then it's not going to work. It's outcomes first. You know, what what outcome are we looking for? And then we use AI for that. then we talked a little bit about the news. And you know, way the ways that companies should kind of keep their ear to the ground in terms of things that are happening because more has happened in my opinion in the last week just with the government getting involved than these are bigger changes than we've seen so far in terms of in terms of government involvement. And I think that's important. ⁓ and then maybe the last thing that we left on the table is just is the real danger. of AI, not that it's too powerful, but that it's too agreeable. Like, does it are are we just building a bus business strategy based on somebody like open AI telling us what we want to hear? I think that's very, very dangerous. And it's another risk that executives really need to start planning for.

Yeah. So I think, you know, ⁓ you know, not to set the conversation up for you know, a follow up directly, but at the same time, you know, th this is the work that you and I are both doing. You know, we're we we've got clients across the country that we're helping with with this type of work, trying to understand s you know, it is the the the old Franklin or ⁓ Stephen Covey ⁓ right, seek first to understand. ⁓ I think

that's where we are as a as a society or as a group of people in business right now. We're trying to understand. There's definitely people that are way out in front, but we're still in a learning mode, very much so.

Yeah. So we're encouraging folk we're encouraging folks to be curious, to ⁓ to seek to understand ⁓ some of the fundamentals of these tools and products. ⁓ we're helping folks so folks can reach out to us and ask us about it. And and we have ⁓ colleagues that are also doing the same that ⁓ but I think you know what we would invite folks to do is is ⁓ you know. Ask us questions about what's going on and what are we seeing as we're talking with ⁓ peers in the industry, ⁓ both who are AI fluent and those who are not. ⁓ and ⁓ if you think you are behind, ⁓ maybe you are, but the the the closing the gap is not that difficult. Yeah. ⁓

Yeah, we are also looking to to go to a few conferences. We'd love to meet up with some folks on, you know, relevant AI conferences that are happening all around the country. There was one in Denver just last week that I didn't get a chance to make it to. I will be speaking in Phoenix in August to a group of finance professionals ⁓ that are trying to get their heads around AI. So I'm looking forward to that. So yeah, lots of learning, lots of engagement. So we'd we'd love for you to be a part of that too.

All right. Well, this has been Veer.

And I am Jim, and we are your that's right.

Functional exam. We gotta

⁓ And I'm Jim and we are the functional executives.

⁓ see you're doing functional.

Okay. I don't know. I mean we never practiced that. We never even talked about it. We do we did that on the spot.

We did. That's right. This is ⁓ well, one of us is ⁓ you know, it's ⁓ it's that ⁓ maybe in future ones we'll ⁓ future talks we'll talk with folks about ⁓ things like that, like timing. Both of us, Jim and I both play music and ⁓ it's one of those things where like you you gotta be you gotta be in the in the pocket to be able to pull that off. We were not.

I don't know. It's probably too cheese. we weren't. No, no, that was more like middle school band than it was like jazz.

All right. Well thanks for it. I appreciate you.