Ryan Staley on AI and the Future of Building AI Workflows in Business

they had the Spurs on there talking about how they implemented you know open Ai and chat GPT there they were like yeah we just implemented it we didn't have any expectations of return and I'm like holy I'm like that's a luxury right I take the opposite approach I look at it through the lens of kind of like two core areas what are the kpis that you need to hit over the next 3 12 months right and let's align the use cases with those kpis or what are the problems like that you're experiencing right now that we could fix immediately and layer with top 1% principles with you know integrated in AI to fix those hi my name is Demitri panichi and I'm a content creator agency owner and AI Enthusiast you're listening to the AI agents podcast brought to you by jot form and featuring

our very own CEO and founder idin tank this is the show where artificial intelligence meets Innovation productivity and the tools shaping the future of work enjoy the show hello everyone and welcome back to this episode of the AI agents podcast in this episode we are talking with Ryan Staley how you doing Ryan awesome man looking forward to being on the show yes we're very excited to have you as well um the connection here has been very cool to see between uh you and uh J for and what you've helped us out with but we would love for everyone to learn a little bit about your background um what you're doing and how um you know this all came to be yeah no problem man I'll try and compress like I don't know 20 7 years of business experience in like a minute I hate it

when I have guests on my podcast that they go on for like 15 minutes straight just on the background so super long story short um I was in the go to Mark I've been in the go to market space for like I said close to 27 years um did the whole kind of stint and the individual contributor side uh from inside sales where I started off or actually saying before that knocking going door too to inside sales to midmarket to Enterprise and then uh also went to the leadership where I was in mid-market leadership Enterprise sales as well and um had multiple PE exits uh as well as built a company helped build a company and that got sold to a publicly traded company so U but one of the things that happened was like I always was asked to basically build teams or create

results with no infrastructure or no basically like team in place or very little resources right but I had massive expect that what needed to be done so it aligns with kind of the bootstrap mentality if you will of just like make it happen and so um in my last stt basically grew an Enterprise Group from zero to 30 million with only five people in oh I'm sorry only four sales people in five years and so after that happened I left and started Consulting and helping companies Implement those systems in their business but about two and a half years ago almost three years ago now I had a guy named Chris Savage from wisti who's the CEO on my podcast which is called the scaleup show adin's been on there just gave some Banger interviews this is now called the AI agent show and effectively he's

like hey Ryan have you heard of this tool called Dolly and I'm like what the hell is Dolly right this is like right when it came out V1 it was terrible right um and I tried it with my daughter and like yeah this is okay but the thing that was really impactful from that one event is I then got access to Chachi PT as soon as it came out and what transformed what I thought was possible specifically for go to market was I tested it on things that took me 10 years to learn and it got 90% of the way there and three questions yeah and so once that was like a sledgehammer like hitting me between the eyes I'm like this is going to change everything and I've been obsessed with uh that ever since and as a result um help companies deploy at

scale uh integrating AI into their people to multiply the capacity profit and revenue without hiring very cool that that was pretty so synct actually you you you did just exactly what you were hoping to um no that's that's awesome I uh I'm very very similar to you on the the AI front where I was introduced to it um it was an interesting experience and then you have some sort of epiphany where you realize it does get a lot of the way there um and the nice thing about small business owners and bootstrapping is the goal usually is to get a lot of the way there not necessarily to get as uh fully there as one could get with employees but part of the way there so um I guess for for you when you're working with companies what is the uh framework you kind of

work in in order to help them scale their teams and I guess what's the first step in order to help somebody scale their teams uh just by implementing something as simple as a large language model in their workflows and workplace yeah it's a great question it well it's it's it's kind of funny because like I'm part of open ai's like private invite only community and they they had the Spurs um on there talking Ai and chat gbt there uh which Spurs the professional basketball team right got it and it was funny like um they were like yeah we just implemented it we didn't have any expectations of return and I'm like holy I'm like like like that's that's a luxury right that's like a business that's making so much money that just like yeah we'll just try it out and see what happens where I

take the opposite approach I look at it through the lens of kind of like two core areas like I know what the patterns are from working with so many individuals and leaders but the the core areas that I have them focused on is either what are the kpis that you need to hit over the next three 12 months right and let's align the use cases with those kpis or what are the problems like that you're experiencing right now that we could fix immediately and layer with like top 1% principles with you know integrated in AI to fix those so that's really the core like framework that I use every single time and if you take it through that mindset there'll be an Roi generated because it's attached to achieving the Enterprise outcomes that you need to to to do so okay cool Rin oh yeah

it's cool uh Ryan how are you doing yeah I'm doing awesome man it's good to see you again as always man I love I love connecting with you man so I'm really appreci being on the show and you have been helping Jon for a long time our team our sales team so uh and we have been meeting with you for a long time and we H we also have like similar interests uh you are also doing the a agents show now it's called air agents show previously it was the scale up show and I was a guest that's how we met initially actually that's how I met Dimitri as well like I was in his show as well uh so and one of the things you do is like when you when you came to our you know working with our team is like you

gave them you know training help them in different aspects of our goto Market efforts and um I'm I'm really curious about like some examples or some case studies uh user stories like obviously you don't have to give names uh you should you know you probably shouldn't but like uh what kind of things really help uh like sales and marketing teams uh what kind of things they they do with AI to really get that uh you know return on investment or more in their efforts like what did you see so far and you have been helping others of companies so you you actually seeing the results so I'm really curious curious about that yeah no it's a great question man and like I have a broader depth of experience uh because I've been working with like companies that are as little as 10 Mill or 15

million all the way till like in the you know billions right so um so what I'm seeing as a pattern well first of all there's like if we're looking at go to market teams I look at this through two lens there's very distinct unique use cases for leadership executive leadership in those areas and then the individual contributors as well so I cut it up into those two different areas because there's certain things you need to do to lead with AI and so there's a lot more strategic executive function capabilities you could leverage the PHD models for and um to give you examples that could be like large scale data analyst or data analysis of looking at like let's say your top your customers let's and here I'll give you a real example for executives that I love so if you look over and this happens

and this consists with every single business I work with is if you look at their client base what you'll see from the deals one over the last 12 months or the last 36 months for LTV or lifetime value if you look over the last 12 months it'll be annual contract value and a lot of times there's weird outliers that exist in the top 20% of clients that if you just double down on that you are going to like quadruple your deal size um without hiring anybody without doing anything different just by sheer focusing on that weird outlier and so what you could do is you could use the data analysis tool in 40 to give you really good insights and patterns on who those people are and how to like Niche down with that and refine your focus so that is a strategy that Vista

Equity uses to turn companies from 1 billion to three billion I use that manually when I was a rep and as it go to market and so that'll double your deal size or quadruple your deal size but also at the same time shorten the sales process so that's one of the use cases I love for executives and then I have more if you want me to on the executive side and then I can get into the side but I'll take a breath in case you have any follow questions on that you me to no I think that that that's um that's interesting um I uh resonate with your point on the difference in how AI can help an individual in just you know your everyday workplace position uh versus like a company operator versus like someone in in in the higher level of positions um

when you're working with companies what do you find is the hardest spot B for Buy in to occur right because obviously it's going to take buy it could be immediately bought in by people in the higher ups but what do you find is the most difficult uh part of getting buying at a company for actual Implement is it implementation of it or because a lot of people conceptually will will be fine with it but uh implementation I'm sure could be an interesting factor in there too you know it's interesting so I think um so the the process I usually is like I'll analyze those problems then I'll do a combination of training and hackathons to implement it sure okay um but then also like really synthesize the people that are that are that are using the tools and providing use cases and really promote the

success that they're doing so I think sometimes like there's still the misnomer of Executives where they're like hey I can just delegate this to my team and do it right like that's one of the biggest gaps because like you need to lead by example an executive like on how to use it and what you're using it for because it's so different than any other SAS tool where it's not just a linear output that it provides yes and at the same time like you have access to basically something that costs a million dollars a day to operate uh and for $20 a month or $30 a month or even free and so I think one of the big thing is leaders Miss to that they think they know how to use it for email or they play around with it so they've checked that box where

they only scratch in the surface and are behind 99% of people if they take that approach so that's what I would say is the biggest kind of Gap that people run into with implementation is there any concerns usually in these companies that the outputs that they're going to have when using the tool are not as good as somebody who is you know in like an entry level position or or otherwise because I I feel like that's some part of the buying out because like like so I survey my my um customers when I bring them on and that is the number one consistent pattern that I've seen of people not getting good outputs and so that's one of the things I hit on the first session is like how to get top 1% outputs so that like the results you get are amazing because once

you do that that unlocks a whole different part of someone's brain of what's possible and that's what really gets things flowing I would say that's one and number two is I find out within a week or two how I could save someone 5 to 10 hours a week um through using the tool and that's the other thing that opens up you know their interest and um coachability is what I would say so those two areas interesting uh I've noticed that with people in the general I don't know T asphere that is not in you know and enthralled with AI they tend to have the in they have the Intrigue and then they they're not really great at prompt engineering or they don't put a lot of time into it so then they'll think that you know their the performance is worse than that of like

a basic level employee so that's that's why I always asked that question I find it this like weird just juxtaposition with they're like oh this could be saving me a lot of time but then they try it for a little bit they don't necessarily put in um the best of inputs and then the outputs aren't good and then they they try to blame it on the AI more than the actual performance that it can have it's it's more so your prompt engineering was just bad um is kind of what I've noticed a little bit um yeah I it's true I got the next question uh so uh yeah I actually wrote a book about automation automate your busy and in the in the book that I talk about how to use Automation and and what I encountered is like when I talk to people about

automation is like the biggest problem is like taking that first step like it's just like you know people are already busy like they're already busy like you know we don't have time to uh improve things we don't have time to automate things like that's the biggest struggle and U so someone um thinking about using like a company thinking about using AI for their um you know sales and marketing where do you think they should take the first uh step like what should be their first step like you know so that they can they can get some you know uh they can give it a start and build some momentum like where would be like the uh you know low hanging fruit for them to get results yeah so I I take the the atomic habits approach to this and this is like the simplest way

and basically one of the things they talk about in the book Atomic habits written by James Clear is basically like attach it to something you already do every day right and and start to formulate a habit like that so easiest way to do that is um think about how much on average the average person uses Google a day right like pretty much everybody uses it every day at some point so instead of spending your half hour a day on Google spend 15 minutes on chatu BT when you do it sure try chat BT first try it on there or I actually like perplexity better than Chachi BT for search um I showed my mother who was 75 years old on um how to use it and she's like this is so much easier than right and that's a 75y old woman who's love my Mom

amazing but not the most technically Savvy person sure and so if a 75-year old grandma can learn how to use an AI tool like like that um when I had lunch with her I showed her how to use it that's it right like then anybody else could do that so if you just replace it you'll find your life getting better and better and then that'll get the momentum going wow the voice mode really helpful in the in the start like you know both for my kids like you know they're in uh Elementary School and for my like father and mother like uh like it's really hard to get some to use technology but uh once I gave them the like the chat GPT voice mode into their hands like they were so like it was so easy for them to use it and my kids

are using it to do their homework to do their research you know my father is using it to ask different questions and learn about different things so uh it kind of replaces uh you know searching Google for them and it's it like the voice mode seems to be really promising in this area yeah go ahead di absolutely I gotta my mom how to do that too yeah cool I got I guess I got to get on the train of attempting to get the elderly people or the people that are more senior than I in my life to to use it I haven't I haven't got on that train yet but I guess I need to um I was thinking the the you know the whole perplexity thing does introduce an interesting wrinkle into the mix when you're probably working with companies do you I mean

is it through Partnerships that you default to chpt um or is there anything else have you ever implemented anything else at a company um and is there any reason yeah I've implemented co-pilot I've Implement um I'll make recommendations and tools like jot form right like um I really look at like what's I don't look at everything as a a nail and chat gbt as the hammer right I really look at the use case cases and so um it's kind of a combination of things man I would say like usually chat gbt is the farthest ahead of everybody um in terms of releases in in in advancements and like we could talk about deep research on that um and then other people are like catching up constantly so that's usually the best default um I don't get paid by them I don't have any affiliate fees

or anything sure um I also personally like Claude I like perplexity for search or some like really crazy agentic use cases we could talk about with perplexity that not a lot of people talk about um as well as deep research and then you know like I even like this one's wild but um I'm I mean adakin knows we've had conversations like I'm a go to market first person and spend a ton of time in large language models I'm not a programmer I'm not an AI engineer like I was creating some stuff on lovable that blew my mind um lovable has grown from Z to 10 million in like three months and that's like agentic building I've used replit as well um but those are some examples so it's really use case dependent is what I would say but by default usually chat jbt is the

the kind of catchall because of the PHD models it has in my opinion I think you you have something that you call the AI model metrix can you talk about a little bit about that yeah sure man so the AI model Matrix like I just made a YouTube video about this because it was one of the things that came up so many times when working with clients so like how do I know what model to use when right good question maybe what I'll do is like I don't know maybe I'll I'll I'll share it um maybe I could share a link in the the chat or something of it um I haven't made it publicly available but even know if people are interested I maybe I could do something like that but but basically what it is is it just breaks down every single model

that's available in chbt right now with like the specialty us cases of what it's good for what it's not good for and then actual real use cases to use it so it ranges from 40 all the way down to deep research and then even 4.5 which is the emotional intelligence model that they released um in in research preview recently as well can you sh can you share some examples from there like you know like what type of task are you know more suitable for what kind of models like I think in that Matrix you have you have like all these different models and you have like different use cases uh examples like that if you can share some examples like that would be I think really cool do you want me to pull it up I know you guys do stuff on we do screen

grabs on here so we more than happy to have you pull it up yep okay let me pull it up real quick I will but what I could talk while I'm doing that I mean I know this stuff um it's hardcore embedded in me so if you look at 40 I kind of look at 40 is like the Swiss army knife so you're going to use that for the bulk of what you need in um in terms of tools I think it's really good like the GPT that open AI made uh data analyst is really good for analyzing data and there's a lot of things built into there sure um that I think are really impressive so I think like for the bulk of what you want to use it for um that's really good for search I think if we start to progress and

we want to get to more uh and there's a lot of text on this screen that I'm about to show you so that's why I kind of had to make it into a matrix so why don't I share this right me share my screen actually hold on a second it's bigger so hopefully everyone can see it and then if we shift to like let's say the PHD models so the PHD models are more along the lines of uh 01 03 mini and 03 mini High let me make this bigger so folks that are looking on YouTube can see a better example of this and that way we don't need ah there we go that's probably way better so like what I would say for like 01 and 03 is like if you look at 01 and 03 these are like the Albert Einstein models so

03 mini high is the best at like deep strategic thinking so think of that as like you could use 40 and this is what's called Model stacking to analyze data in like an Excel file get the output or the highlevel understand understanding of it and summary of insights and then you could put it in 03 mini High you could paste those in there put them 03 mini high and turn it from insights to strategy down to Tactical execution and so that's one of the things that i' like to do is stack the models from there um if we look at 4.5 this is the one that's only on limited research and basically what 4.5 is is it's the highest emotional intelligence model so far um there's mixed reviews on it yeah what does it mean yeah so what I mean so based on the way

it's trained it's supposed to have the best understanding of like how people feel and the emotions behind writing and logic and I'm sorry not logic in decision- making 03 and 01 are more logic and stem heavy models so science and math and think of this is more like tapping into how people really feel so that they can write deep rich emotional content now here's the thing I'm going to tell you about that's really important for every single person listening right now this is like a wrer downer 4.5 what's happening with that is effectively when we jump from the normal models to 4.5 there's a 10x Improvement in emotional intelligence with the way this model operates now it's still buggy right now so it's not perfect but GPT fibo is supposed to be released the next two two and a half months and there's supposed to

be another 10x leap from the the core models that we have now to five so 2 10x leaps equals 100x Improvement in that and that's going to basically have the ability to replace a lot of folks that have high emotional quotient like sales or other areas that's involved due to what these models are going to have in GPT 5 and nobody's really thinking about it that's yeah that's big and then last but not least definitely is deep research so we could Riff on some use cases but um some things that companies have used it for that I've worked with are like service addressable market so like what like if they're trying to look at a vertical or a market segment that they're in you know what's available in there what are the gaps what are the opportunities um does a really good job at that

um another example that someone used it is they wanted to move up market and so they were trying to get really good with messaging and understanding of the different verticals so it did a really deep research report on that and then um one of my absolute favorites is um with with that model I used it um for my own business and I'm like hey like there's B basically you get 120 deep research uses a month for the pro user which is a $200 a month user I have access to that um if you have the normal version like teams or plus you only have 10 uses a month so I said okay I have 120 uses of deep research per month how should I use these for my business to grow my business to multiply my clients and do that right and it gave me

120 ideas nice all categorized and I went through that man and I started to like tactically pick on different ones and some of them are really freaking good um so it ranges all the way from market research competitive research messaging um prospecting all those kind of different areas now we only need the Deep work uh model which will actually do all those yeah yeah yeah it would be it would be based off Kell newport's wonderful book yes yeah that'd be funny exactly uh I guess my follow-up question to that would be you know talking about deep research um I have used CH gpt's deep research I think it's awesome um there is a lot of great stuff I'm seeing when when I try it out I do have the team plan so I only have you know so many but so many uses per month

but I tried to be as effective as I can with them however there are other tools that have deep research now I believe it's perplexity I believe it's also not only perplexity but Gemini as well didn't they they just release something um on on that realm what are your thoughts on those other models and how their deep research is and how they compare yeah so there's Gemini perplexity grock and Claude just released it as well okay so here's what I've noticed like I think I've tested all of them and because like I look at this like I said I have no Alliance to any model I I like to just use what works best that's why I use a bunch of different models right I'm working on something for that because I think that's an opportunity um but what I would say is like one

of the things here let me give you example like instead of just like being negative yeah I'm not going to be the opposite side where I'm going to be like a classic YouTuber like this is amazing this is the best thing that's ever happened right um but I'll give you a good use case in perplexity that I've never heard anyone else talk about and so this might be valuable so with imp perplexity and you could go in there and basically click on the mode that you want maybe I'll pull this up and I'll show this as well but one of the cool things about perplexity is like you could use uh deep seek as an example um is one of the models I have the pro model which is basically uh the $20 a month version so let me share this again and I'll show

you because not a lot of people know this and this might be a huge level up okay all right okay so if we click on here one of the things is you could use the different reason models so um this is kind of like and this is what I noticed uh one of the interesting things about perplexities they tip they test open eyes Advanced models before the public because I've seen that but basically there's probe there's uh reasoning and then there's deep research for perplexity deep research I was not impressed it kept timing out it like complete job but this basically has the reasoning model is effectively like deep research and I could use any of the models here so that's something that's really cool the part that nobody really I've heard anybody talk about is this button right so you could set this sources

for search so you could look at Social academic or web it's it's defaulted to web now here's where it gets really cool if I were to do this let's say for only academic I could type in here and say okay identify the the top 10 papers on um let's say LM prompting for agentic use case optimization I'm just making this up on the fly right but what it's going to do is it's only going to look at research papers versus pulling everything it can off the web and so there's a different level of quality you're going to get from that in terms of doing it and as you could see even though it's not deep research it's mimicking that it's going through the the kind of like Chain of Thought with it and what you're going to see is it works pretty good and pretty

fast so like this is what I would say is um an example of how you could use it and it didn't pull up 10 papers ironically but it's because it says there's only one or two one paper on this which is really interesting um so this might be and then it gives you a link to the research paper as well so um just a little bit of a different way that you you could use it um and then at the same time like if you're trying to get sentiment off of like what your buyers are doing you could put that and put social media and have only responses from Reddit so like for example like we're looking at doing research for job form I could say okay what are all the agents for um customer support that would be needed specifically for nonprofits or specifically

for Education institutions that are talked about in Reddit and then give me examples of what forums are so it's really good for things like that as well okay so here's the things that I'm noticing every product seems to have uh a unique wrinkle somewhere lying within it you know um however there's a lot of different services that you'd end up needing to purchase which I have no problem with I mean we we run a podcast about AI I make it I have a Content agency it just makes sense what is your elevator pitch to the average person um in a company small business or otherwise for why you'd want to invest in in in not only one tool but multiple tools uh CU I personally can see the value in having in spending 40 60 100 whatever it is a month on AI and automation

but what what's what's your elevator pitch for that yeah I mean as a simplest like I think you pick one large language model OKAY standard I for your company and then you could stack the other ones for Niche use cases right so the the cool thing about perplexity is it's free like you don't even need to pay for it and it's it's research in real-time research is like massively good for Time Savings alone so like even if you just had like chat chbt um Gemini is finally catching up it's not quite there yet but it's getting better from Google like they're making huge advancements um and claude's really good too like so cool thing is now is like unless you're a super ridiculous power user like me and have to provide like always be the curve for your clients you don't necessarily need to have

the exact power up for every single one of these sure it's going to be more use case dependent right so I think here's what I say my two cents standardize on a platform have your team work together collaborate and share and then for those Niche outlier use cases you might have people or departments that have multiple licenses right that's probably the best way to do it okay cool and the next question I have in in this sequence would be you know we're talking about using AI in real time right we're talking about using AI in uh what I guess I would think of as version one of how you could probably interact with with AI and in in my perspective there's like kind of three tiers you can work at it with there's um just chat you know using it in the day-to-day interacting with

the computer there's number two which is making automation sequences and then there's number three which is uh agentic um uses that are also autonomous so how do you at a company and have you worked in companies where you've helped them Implement uh Automation workflows and like how do you get them to take the step from just the normal day-to-day chat interaction and also use it for what in my perspective is actually even more practical more effective more time-saving which is like automation stuff on top of uh just interacting with the chat yeah that was a great question man' I think here's what I would say this surprises me a lot of companies don't even know understand how to use the basic chat that's yeah like we're talking companies with hundreds of millions of dollars in Revenue no freaking clue there's there's no standardized approach for

their employees so they'll be selling AI products but they have no AI strategy for their team and there's been no training for their team and so that's problematic because then people operate in the shadows and they won't share their Innovations organizationally and that's one of the biggest things like you could be sitting on a gold m of innovation but if you don't put that training out in the public then that's not going to happen so I think that's the activation point and once they realize the employee realiz it's encourage employers behind them then the floodgates open of opportunities so that's going to give you your best um automation use cases um however I would say there's a big difference of automation between Enterprise companies or small and medium businesses and small and medium businesses I mean you know 50 people are less right big big

difference um but I think it's a natural progression because what happens is once people are activated they understand the use case to the benefit they're like all right cool next step is like how can we connect these to a couple different systems right so it's a natural like next step that they have because they don't want the constant clicking right or they don't want to copy and paste in a lot of different fields so that leads to more the automation component the only thing I would say to with Automation and agentic is it's it's still Limited in scope right now um and what I mean is like you can't just say like I want you to create an outbound campaign and you know generate a thousand leads for me no like it's it's not there so you got to be very specific and have agents

that are defined for narrow tasks and if you do that and give them the right training data then they could be really really good at execution because like it doesn't matter if you could automate everything at scale if you're just automating shitty processes you know what I mean you're just automating slot and that creates more problems than good so that's my two you're Auto automating it's you you pref do you prefer uh just taking something that they already do and then like start using AI for various parts of it and just you know doing the same thing but you know with the help of AI and maybe scaling it with the help of AI or do you prefer like starting from scratch hey uh let's just start from scratch and let's let's build this new automation like which one uh do you prefer when you're

working with company well here's the thing I mean most of like the automation stuff like it's just like normal automation with like some AI reasoning on top of it you know what I mean it's not like so like I think that's where a lot of people misunderstand it you know um and then like if we're talking about agents like I've seen a lot of people try I've tested a massive amount of tools and a lot of it's not that good like like I like I could say like I could put in a prompt and it's better than the output where it's got like 17 agents talking to each other and doing everything right so like I would say like Simplicity over complexity is is going to like help you nail execution and but but here's what I would say there's companies like jform that are

like starting to democratize what is available with agents to make it really easy for people to do that and as the technology gets better and better then that Nirvana of like using deep research is going to be available across a magnitude of different use cases and then that's going to explode so that's my two sents what do you think about the future like like uh obviously the AI agents uh they do work but it's not like you know uh like you have to really spend a lot of time on them to fix them and get them to a point where they are you know useful um and you can actually make them useful but it it's actually requires a lot of effort and you know skills as well um but when you look at like five or 10 year Horizon like where do you see

us uh and the companies is like you know what do you think is going to happen in the future with AI agent and in terms of sales and marketing especially so let's say I'm just looking at my my date so it's March of 20 25 right now so let's let's take five years because I think if I think if we look at 10 years people's heads are going to explode right yeah agree so if we took a look at five I mean the this is crazy man so it's a great question anak and it's like I started thinking about this a lot over the weekend I watched the movie electric state did you guys ever watch watch that or see that just came out it it's gotten a lot of like it's with Chris Pratt and uh was it Millie Bobby Brown from yeah from

stranger things stranger things right so it didn't get great reviews I thought it was entertaining it kind of reminded me of Ready Player one where there's like an alternate reality that happened in the 90s because robots basically took over and there's this whole big thing right so so it was kind of funny because it was like stuck in the 90s right so that got me thinking and I started I started asking deep research like all right I think we're going to hit artificial general intelligence in the next couple years which is basically where AI could do all the tasks that humans could do at the same or higher level um and the chief product officer of of like open ey thinks that that could be hit by the end of 26 I think or 26 um Dario amade from anthropics said 27 so like that's

gonna massively change everything and so um my take is like there's nothing that's happened faster in our life from a technology change than what's happening now we're at Moors law times three and so I guess like the biggest thing is like there's going to be so many industries that are reinvented um so many jobs that are reinvented and there's going to be like I don't know man like I ran out 20 different scenarios that deep research came and it went from like a surveillance government like Run State all the way towards more the utop model it's probably going to be somewhere in between but I think government regulation is going to be one of the top priorities and the reason being is because there's an arms race between the US and China right now yeah um because there's a a security issue where if someone's

AI is so good they could literally attack the entire infrastructure of another company like a nuclear bomb and and destroy it and so there's a lot riding on this and so long story short if we're looking at life I think it's going to be unrecognizable it's going to be the difference between today and like what happened in the 1500s probably 3 four years from now um which is crazy and then government is going to have a big role in it um and at the same time um I do think there's going to be a lot of upside too because people were at the largest unengaged rate of work in 10 years really and people are ready for a change and I think this will give us an opportunity to make some areas of work really fun and really interesting and exciting as well that's great

yeah yeah it's a good yeah it's a lot it's a lot especially with the um stuff on the you know the the state level or well country level uh side of things and and what's working behind the scenes there that we're not all privy to but we probably should assume is going on um I was thinking about it though when you're when you're talking about this it's it's baffling to me a where we'll in 5 to 10 years um especially with the the workplace you know um I was talking with a buddy of mine about agentic AI the other day um even just the concept of training in Ai and a knowledge base a Persona these types of things that your average person isn't familiar with you know hyperfocused use cases can be really really well produced now um what are your thoughts on where

this will kind of impact the entry level positions at a lot of knowledge um knowledge work companies cuz I do do have some I have a feeling that AI agents are going to take over a lot of those roles within the next five years um in some respect I don't know whether the roles will be gone or whether a lot of work will be um just offloaded from them what are your thoughts on that you know that's definitely a possibility the the unique thing though like and I didn't start thinking about this until um adakin mentioned you know his kids were using it for voice and his parents like I love that like I got to start using that for my parents as well because you know typing is hard um I like I have the you would think I have like fingers that are

10 times the size because of how much I fat finger stuff on the keyboard I suck at typing right but um so I lost my train of thought what was the question the question is like the question is in uh the future of the workplace do you feel like this AI agent stuff will Associates I'm back you got me back Demetri I'm back man okay so the future of the work so basically this is the first time ever that we're talking about like where could create the future of the workplace the thing that nobody's really factoring in is we have kids in high school yeah that might be more ad dep at using AI than uh people in the workforce right now that's fair and I don't think any anybody's really thinking about that so you're like yeah the like the entry level jobs are

going to go away yeah I think the entry-level jobs as today are going to go away but if you have kids that are like have been crushing it with using AI for their whole life and you have some folks that have a ton of experience but don't they're going to get clown by these kids um and it's something like I never really thought about we talk about right now because like they are going to operate a superhuman level at 10 times the speed um that some people that have way more experience than man so there could be big opportunities for them to jump faster and further and work for well you know what I mean good point we all have to retire then you know yeah I mean this is I don't I don't know how well familiar with um there's a piece of work

from I think think 19 I want to say the 1920s by Bertrand Russell it's called impraise of idleness it's actually it's kind of a philosophical uh paper he posited a point it was post World War I and he uh came to the Epiphany that essentially the entirety of Europe you know so Mo modern world you know as he put it um essentially had half of its work floors arbitrarily blowing each other up for for you know 6 years and then they just went right back to work and they're like he's like wait a second and the economy was fine uh he then posited the the point that work in the next few years and he's he was way off obviously has to essentially change there is going to be an adjustment with the Advent of more he actually used the word automation but from the

industrial terms right so and I he was saying automation he was saying factories these types of things he was saying work has to change from a frequency standpoint now it's hilarious because as we know in American culture that did not happen we just kept working uh same with a lot of the world maybe in Europe a little bit less you know France is down to I think like a 30-hour work week or something like that at this point which um in America that's not the that's not the vibe at all that I'm getting um of what's to change but do you think that practically with what burin said actually a hundred years ago at this point with the Advent of AI we actually might start seeing the first real regression in uh a culture change in work because we've had the probably the ability to

slow down a little bit we just as I think as a country we just choose to like not because growth or something yeah I mean it's a great question man it's something to to Think Through it's like unless it's and let me put it this way I think if if PE people would definitely take that route like you saw that with work work from home that's why a lot of companies want people to stop working yep working 20 hours a week or whatever so like I think on the employee side they would love that I think on the owner side they would not love that because like you know they look at that as unused capacity from people and however if if people understand that they can get replaced if if they don't modify by how they're working then that might change the mindset a

little bit and I I've seen people that work hard too but um yeah I mean it's a really really good question I think eventually the technology will get so good that that'll be a requirement because there'll be a lot less work to do um at least for the majority there's still going to be hard charging companies that have massive expectations of course um always but I think it it could create a new it's just like how the work from home created a new paradigm of how everyone works I think this will do the same thing but just in a different way yeah we'll see uh you familiar with the um it's a trend where people essentially secretly have multiple jobs I'm forgetting the name of the quiet quitting not quiet quitting it's the uh overemployed have you heard the overemployed movement yeah I've seen people

I've heard of people doing people are going to start doing the AI agents overemployed uh yeah you're going to sh I mean um it's gonna get wild and we gotta be careful um because like at the end of the day like I think it's good to have integrity with what you're doing 100% say what you do and do what you say and um you know if people operate that way um sometimes it's going to make sense depending on what your job is and your commitments are but at the same time sometimes it's like it's not the way that I would want to be treated um so like that's going to catch up with people I think if they do that but it's it's something that's stuff we'll be out there so something to think about it was just some Food For Thoughts stuff before we

close things out we only have a couple minutes but the last question I do have for you actually in this realm is more so in regards to AI that has the ability to do stuff uh for you in more than just a chat ecosystem or an automation ecosystem I'm curious what your thoughts are on these new operator and computer-based uh tools I know we we just did an episode on Manis which was very intriguing um I was mainly baffled by the speed at which it operated what are your thoughts on that and have you uh seen anybody implement it in an interesting way so far I know that's that one's in beta but operator probably more so I love the potential um I haven't used Manis I think one of the flaws that folks have talked about with Manis is like it can't use any

signin right so can't sign in your Google Sheets and create that I can't sign in your LinkedIn account I can't um it supposedly gives you access where they don't see that I you never know with that stuff right whether or not they're monitoring that in the same window um I tried it with operator I tried it with convergence I did a episode on that and um I was really impressed with I think it's convergence convergence converges um It's a UK company that that basically is doing what what operator is supposed to do but um it kept crashing right and it worked really slow however you use deep research and it'll do the same thing without you having to Monitor and check it so like I spent hours using operator trying to get one task done um so here's my point my long story short of

this just like I gave with my 27 years of career in in two minutes is um I think I'm really excited about it I love the concept of what man where they connected Claude 3.7 to 27 different tools to operate at a different level I think that's amazing Innovation I just think it's like the early days when we're using 3.5 or four where you'd have to prompt like an absolute genius to get a good result and I think that's kind of where we're at with some of this operator computer use stuff right now however by the end of this year I I see it being completely different because people are going to double down Triple Town because they see the utility in it and it's going to change things massively very cool by anything last from you yeah the last question is where can our

audience find you like uh how to how to find your podcast and uh your website and you how can they reach you yeah yeah man um so you could my website is ryan. I think like the best way to to get connected to me is on LinkedIn send me a DM message there um I publish AI content every day my journey what I'm seeing work well not work as well as the latest and then um you know if you like content like this my podcast in um YouTube where I show actually explainers on this is it my my YouTube channel which is Ryan Staley so um so that's pretty much it man LinkedIn DM me men sure you saw me on the show YouTube if you want to watch some some great content would love to hear from you especially if you want to implement

this at your company and transform your entire team in less than two months so that's awesome it's funny it It's it's uh when you said when ID and asked the last question um have you seen ioot yeah it was funny I thought so for one both of us apparently are from the Chicago land area and that movie takes place in Chicago and I was going to say to idin oh that's the right question to ask at the end of it and if you recall the the scientist always tells Will Smith when he's in the Hologram that is the right question I love I love that movie I watch all right guys thank you so much for listening to this one this was a fun please check out all of Ryan stuff leave us a review on Apple podcast Spotify any of the different platforms that

you're watching follow us for more thank you so much for listening and we'll see you in the next episode bye