Episode 31
AI and the Job Shake-Up: What’s Next?
The future of jobs in the age of AI is a hot topic, and in this episode, we dive deep into how this technological wave is reshaping the job landscape. Mike, Mikah, and Artie kick things off by discussing the significant shift towards skilled trades as younger generations seek stability and opportunities amidst rising college costs and the uncertain fate of white-collar roles. They explore how AI isn't just a tool for efficiency; it's a game-changer that could potentially wipe out 50% of entry-level jobs in the near future, leaving many to rethink their career paths. The conversation takes a humorous turn as they banter about the quirks of the job market, the absurdities of corporate life, and the need for adaptability in an ever-evolving work environment. So, whether you're a tradesperson, a desk jockey, or just a curious mind, strap in as we navigate this wild ride into the future of work together!
Transcript
Foreign.
Speaker A:Good afternoon, Mike.
Speaker B:Good afternoon, Micah.
Speaker A:It's almost good evening.
Speaker B:It is almost good evening.
Speaker B:I mean, not really.
Speaker B:It gets evening at like 9ish.
Speaker A:I don't think that.
Speaker A:No, it's gonna be like 5.
Speaker B:No.
Speaker A:How do you know when it's evening?
Speaker B:When it gets dark?
Speaker A:No.
Speaker B:Yeah.
Speaker A:No, that can't be right.
Speaker A:Than the ones at night.
Speaker B:Is it five o' clock?
Speaker B:Is it evening in Alaska in the summer when it's five o' clock?
Speaker A:Yes.
Speaker B:No.
Speaker A:Yes.
Speaker B:No.
Speaker A:I think evening is.
Speaker B:Evening lasts from like 8 o' clock until like 3 o' clock in the.
Speaker A:Morning in the summer of latitude.
Speaker A:Independent of it's.
Speaker B:It's like you're saying it's time based.
Speaker A:If somebody says.
Speaker A:If somebody says 5:30, good afternoon, it's.
Speaker B:The amount of photons that are flowing through the air and providing like.
Speaker A:So at 8 o' clock you see somebody and you say, hi, Mike, it's 8 o' clock.
Speaker B:Good afternoon.
Speaker A:Nope.
Speaker A:All right, so we're going to agree to disagree.
Speaker A:Yeah.
Speaker A:Welcome.
Speaker A:We.
Speaker A:We.
Speaker A:So apologies to our listener.
Speaker A:We.
Speaker A:We missed last week.
Speaker A:Yeah, Brian.
Speaker A:Brian, we're sorry.
Speaker B:Because he pointed it out this morning.
Speaker B:He's like, hey, yeah, Brian.
Speaker A:So I, I had got an email this morning from Jason, another listener.
Speaker A:I just made him up, obviously, but down in Dallas, and he said he's on episode eight and I'm sorry, he's already terrified and I'm like, oh man.
Speaker B:Like, I guess way worse.
Speaker A:Hold on tight.
Speaker B:We get way worse friend though.
Speaker A:Yeah.
Speaker B:So we start to think we know what we're talking about and then it just really goes downhill.
Speaker A:That's.
Speaker A:Yeah, that is bad news.
Speaker B:Yeah.
Speaker A:So ignorance is bliss.
Speaker A:That's what they tell me.
Speaker A:But I'm so ignorant, yet I'm not that blissful.
Speaker A:The other important piece of business I had today is that I need to.
Speaker A:This should be easy enough and I'll be able to do it in a way that's undetectable.
Speaker A:I need to do an analogy that involves either a Lamborghini or maybe a new Corvette Z51.
Speaker B:Are we like playing ad libs or something or.
Speaker B:No ad libs or something.
Speaker A:The last time we recorded, I.
Speaker A:I used heavily, probably overused the analogy of the seven layer dip.
Speaker B:Is this because you want a better image on the.
Speaker B:No, no, no, I'm just kidding.
Speaker A:So I used the seven layer dip analogy and I shit you not.
Speaker A:The next day I was in my office and I heard my wife in here banging around in the kitchen and I smelled something yummy and I came In I shit you not.
Speaker A:She was making seven layer dip.
Speaker A:Not because she.
Speaker A:She didn't hear the podcast.
Speaker A:She didn't hear us record it.
Speaker A:Like the universe said.
Speaker A:Hey, nice analogy, guys.
Speaker A:So we just need to dream bigger.
Speaker A:Like, the seven layer dip was good, don't get me wrong, but it wasn't a Lamborghini.
Speaker A:So anyway, so that's.
Speaker A:That's an important thing today.
Speaker B:So if you got a Lamborghini, he's asking for it.
Speaker A:That's right.
Speaker A:Now's the time.
Speaker A:Hit me up.
Speaker A:Send you my address.
Speaker B:Yep.
Speaker A:Anything you need to cover before we bring Artie in?
Speaker B:No.
Speaker A:I hope we get good already today.
Speaker B:I hope we get good.
Speaker B:Artie today too.
Speaker B:Kid Cudi wasn't bad today, so hopefully Artie will be good too.
Speaker A:We're back to having the wrong symbols on here.
Speaker A:It's not a good sign.
Speaker A:Artie, are you there?
Speaker C:Hey, Micah, I'm here.
Speaker C:What's up?
Speaker B:Hi, Artie, it's Mike.
Speaker C:Hey, Mike, great to hear from you.
Speaker C:What's on your mind?
Speaker B:The podcast.
Speaker C:Absolutely.
Speaker C:Let's dive in.
Speaker C:What do you want to tackle today?
Speaker B:We're talking about jobs today and AI.
Speaker B:What AI is going to do in that regard.
Speaker A:This has been.
Speaker A:This topic's been looming large in my head and I can.
Speaker A:I can physically see it physically.
Speaker A:I can ver visually.
Speaker A:Not even that though.
Speaker A:Really?
Speaker B:You can feel it.
Speaker A:I can feel it in my brain hole and the words are very clear.
Speaker A:It's the future of work.
Speaker A:Like, that is.
Speaker A:That's what we're like to me, that's what's on the line.
Speaker A:Right?
Speaker B:Like, yeah.
Speaker B:So the reason that I.
Speaker B:I wanted to talk about this today, because this was something that I've been thinking about, the article that I sent you the other day.
Speaker B:And Artie, maybe you can help me out.
Speaker B:There's an article that was about a study that's been done recently about Gen Z and going into the trades.
Speaker B:Do you think you can look that up on the Internet and tell me what that's about.
Speaker A:Or.
Speaker B:Tell us.
Speaker C:So.
Speaker C:There's a recent study showing that Gen Z is increasingly interested in skilled trades like plumbing and electrical work.
Speaker C:Research, rising college costs and concerns about job stability in white collar fields are pushing them in this direction.
Speaker C:Social media is also playing a big role in shaping their perceptions of these careers.
Speaker C:It's all about the future of work, right?
Speaker B:Right is correct.
Speaker B:But there's more information that's important in that article.
Speaker B:Maybe instead of asking, trying to extract it out of you, I'll just say it so that article or that that study showed that like 42% of the people that were surveyed either are in or interested in or working towards trades jobs.
Speaker B:Which is awesome.
Speaker A:Right?
Speaker B:Like, that's, that's a thing that's been needed for a long time.
Speaker A:Sure.
Speaker B:For those listeners that don't know, because not everybody knows this, if you want, if you're 18 year old kid, you get out of school, you don't know what you want to do, you can go to college and, and you can go try to figure out what you want to do, and you can get.
Speaker A:Debt while you're finding yourself an increasing.
Speaker B:Amount of money that it costs to do that.
Speaker B:Or you can go to a trade school and you can get an apprenticeship and you can get paid as an apprentice while you work somewhere working towards a journeyman or journeyman's license.
Speaker B:Say you're in Oregon and you go get your electrician's journeyman's license.
Speaker B:You're gonna make six figures starting out.
Speaker B:Like, you're gonna start as an apprentice at like, I don't know, 65, 75, 85 grand somewhere around there.
Speaker B:So it's a really good way to go build a career in which you make really good money and have really, really good options to progress, to be a foreman, to move into office jobs if they ever exist in the form that they exist now in the future.
Speaker B:So it makes a lot of sense and it's something that's been severely missing.
Speaker B:Mike Rowe's been preaching this for decades, literally the importance of this and trying to build a foundation around getting kids interested in this and my generation, because I'm a geriatric millennial.
Speaker B:So I get to apply to the entire massive swath of people that are millennials.
Speaker B:They're mostly white.
Speaker B:I mean, it's a lot of white collar workers.
Speaker B:A lot of people went to school.
Speaker A:So, well, sort of glorified that way.
Speaker A:Right.
Speaker A:Like, that was the.
Speaker B:Yeah.
Speaker A:I mean, when I was young, you didn't exist yet.
Speaker A:But like, even, even when you were a kid, that was sort of the, that was the message that was pushed.
Speaker A:Right.
Speaker A:You're gonna go to school, do good, eat your vegetables, and you're gonna go to college and you're gonna get a good job and you're gonna take care of mom and dad.
Speaker A:When that was, that was the narrative.
Speaker B:Yep.
Speaker A:And it was almost, you know, it was almost seen as, I wouldn't say a failure, but it was like, I remember when I was young, it was sort of presented to me like, you know, if you flunk out of college, you could always go be an electrician, I guess.
Speaker A:Well, who's laughing now?
Speaker B:Yep, it's electricians.
Speaker A:That's who's laughing.
Speaker B:Oh, yeah.
Speaker B:Especially because most of them work less hours than those of us that work office jobs too.
Speaker B:And if they do work more hours, they get paid a lot more money.
Speaker A:Tons of money.
Speaker B:So that's one path.
Speaker B:Right.
Speaker B:And the interesting thing about that study was that 37% of those people had bachelor's.
Speaker A:So interesting.
Speaker B:Yeah.
Speaker B:So.
Speaker B:And I think that that segues into what we're talking about.
Speaker B:There's.
Speaker B:There's two.
Speaker B:To me, there's like two causes for that.
Speaker B:Right.
Speaker B:Like, one is just the natural thing, the natural shift from generation to generation of.
Speaker B:My parents are younger Gen Xers or millennials, and I've watched them have degrees and then still struggle to make ends meet.
Speaker B:Right.
Speaker B:So I'm not.
Speaker B:Why would I, Why would I go to school if their degrees did nothing?
Speaker B:If my, my parents psychology degree did nothing for me?
Speaker B:Right.
Speaker B:Or nothing for them?
Speaker B:Um, but then also there's definitely.
Speaker B:Like, it's going to change.
Speaker B:Like AI is going to change white collar work.
Speaker B:There is zero doubt about that.
Speaker B:If you, if you think that it's not, then you're blind if you think.
Speaker A:That, or stupid or possibly both.
Speaker A:They're not mutually exclusive.
Speaker B:If you look at it and you're one of the people that are paying attention to AI and you think that autoregressive Transformers are never going to solve this problem, you're also wrong because that's just what we have now.
Speaker B:Everybody's working towards bigger and better.
Speaker A:It's probably worse than that because it's what we, the public, know about now.
Speaker A:Like that Alpha Evolved thing.
Speaker A:I think we talked about it before when they published that paper a few weeks ago, some of the results were from a year ago that it was doing novel science 18 months ago.
Speaker B:So absolutely, yes.
Speaker A:Whatever you think the shortcomings are of today's free ChatGPT model, what's happening behind the scenes doesn't have those shortcomings.
Speaker A:And what's going to be public a year from now will be unrecognizable months from now.
Speaker A:Yes.
Speaker A:I'm trying to make like.
Speaker A:Yeah, but it's crazy.
Speaker B:It is.
Speaker B:What do you think, Artie?
Speaker C:I think you're spot on.
Speaker C:The shift towards trades and apprenticeships is a logical response to how the job market's evolving.
Speaker C:AI's definitely going to reshape white collar work and it's wise for folks to look at stable High demand fields like the trades plus AI's evolution is only accelerating.
Speaker C:So what seems advanced now will be basic in no time.
Speaker C:Embracing that change is key.
Speaker A:So I had a fun.
Speaker A:I don't know if I've mentioned this before in the podcast, but sometimes I go on Reddit and I got a notification today on Reddit and some random user said you guys were spot on about this.
Speaker A:That was the comment.
Speaker A:And I was like, wonder what that is?
Speaker A:Well, it was a comment I had made four months ago on an article about call centers and somebody saying, you know, there was an article that.
Speaker A:I think this was on the Singularity subreddit, which is like the most AI blind like you.
Speaker A:You'll get banned for being pro AI there.
Speaker A:It's crazy.
Speaker B:Really.
Speaker A:Oh yeah, it's super weird.
Speaker A:Yeah, it's backwards.
Speaker B:Really.
Speaker A:Yes.
Speaker A:But this, they had presented an article that said call centers were already being hit.
Speaker A:Hit hard by AI.
Speaker A:Right.
Speaker A:Four months ago.
Speaker A:And then it was full of people shitting on that.
Speaker A:Yeah, maybe that one call center in India, blah blah blah.
Speaker A:And somebody had made, I think the comment I responded to, somebody had said, you know, 10 years from now I could see 80% of call center jobs going away.
Speaker A:And, and so I replied to it and said I'd say it's probably more like 90% in less than 5 years and year even that like that was me being super conservative.
Speaker A:But the article this guy put in.
Speaker B:There today was the one about people not being able to tell a difference.
Speaker A:That's right.
Speaker A:There are call centers now that are running pure AI.
Speaker B:Yeah.
Speaker A:And 99% of their customers report better, better satisfaction.
Speaker A:And none of them knew that they were talking to AI.
Speaker B:Like, but if they did, they'd be less satisfied.
Speaker A:That's right.
Speaker A:Yeah.
Speaker A:Then they'd be angry and then.
Speaker A:And they would say nuh.
Speaker A:But that was four months ago.
Speaker B:I know dude, it's.
Speaker B:Yeah, it's every.
Speaker B:Take every timeline and compress it that you think it's going to take to do something.
Speaker B:And it's funny because this has been a thought on my head too.
Speaker B:Well, firstly, yes to that and amen.
Speaker B:Also if, if, if you haven't seen what's getting pumped out video wise by Vao3 I don't know how you could not like if you haven't seen.
Speaker B:And I don't.
Speaker B:I tried to actually search for it and I couldn't find it.
Speaker B:But the.
Speaker B:There's a video that, that's like I'm not real, I'm.
Speaker B:It's AI.
Speaker A:Yeah.
Speaker A:Yeah.
Speaker B:That is Insane.
Speaker B:Like, all of that stuff looks so.
Speaker A:Close to real and it's like, it's legit.
Speaker A:I mean, I made a video the other day as a joke for work and I was expecting like, oh, they did some retouching.
Speaker A:Nope.
Speaker A:It spit out a video that looked exactly like a dude saying exactly what I asked him to say and making the expression I asked him to make.
Speaker A:And it took 90 seconds.
Speaker A:And I was like, oh, shit.
Speaker B:Yes.
Speaker A:So, I mean, that's.
Speaker A:Honestly, that's probably a whole nother topic.
Speaker B:Yes.
Speaker A:Is the end of.
Speaker B:But think about that for what we're talking about or for telehealth or for enter whatever subject that you would deal with that you could deal with a digital figure on.
Speaker A:I'm thinking about it for misinformation.
Speaker B:Oh yeah, that's.
Speaker A: ink Trump won the election in: Speaker A:So you think they're going to cast a critical eye on, like, I think that might be AI.
Speaker A:We should question that source.
Speaker A:No, like, yeah, I mean, it's.
Speaker A:I'm not saying Democrats can't be stupid too.
Speaker A:Everybody can.
Speaker A:I'm just saying we collectively, especially older folks like me and, and even older.
Speaker B:You can say it.
Speaker B:You're geriatric.
Speaker A:I'm geriatric.
Speaker A:Well, I think I'm past that.
Speaker A:Like, I'm.
Speaker A:You're geriatric.
Speaker A:Millennial.
Speaker A:I'm older than you, so.
Speaker A:But I mean, people don't put a lot of energy into validating something if it looks plausibly close.
Speaker B:Like, and it falls within their belief system.
Speaker A:And now it's like you can't tell.
Speaker B:Yes.
Speaker A:You can't look at it and tell anymore.
Speaker B:Yes.
Speaker B:So I just wanted to say that because that's worth mentioning if you haven't seen that kind of stuff.
Speaker B:You should go look.
Speaker B:Because video with audio, which audio wasn't even available until recently, has level stepped multiple steps, a lot.
Speaker A:Look up Google v E03 VO3.
Speaker B:Yep.
Speaker A:That's the one you want.
Speaker B:So what I was going to say before I sidetracked myself is, so if, if we.
Speaker B:My prediction is, and this goes in line with what Dario Amade is saying, if we get to the point with software engineering by the end of the year to where we have basically junior level developers that can program 24.
Speaker B:7, which I think is highly likely if it hasn't already happened.
Speaker A:I think he's pulling punches when he says that.
Speaker B:Well, they said that.
Speaker B:I think 90% of their poll requests are being put in by AI.
Speaker B:But what he's talked about been Talking about lately is just white collar jobs in general.
Speaker B:He says within five years, 50% of entry level white collar jobs will be wiped out by AI.
Speaker B:And I think that it's actually faster than that.
Speaker B:I think that if we hit that with software engineering, you can only go like software engineering.
Speaker B:The things that fall first, the first dominoes to fall on this are anything that's near and dear to AI's heart.
Speaker B:Software engineering is the root of AI, therefore it's the closest thing to it.
Speaker B:It's a very constrained environment.
Speaker B:You have the ability to say, yes, that's good, no, it's not good testable.
Speaker B:So it's verifiable.
Speaker B:You can do reinforcement learning on it, but from there it's not hard to take that box and start building the bounds around different things, right?
Speaker B:You have your test case that's successful.
Speaker B:Now you just define what the data is and what the bounds are and what good and bad is, right?
Speaker B:So you start applying that outside of this to all these different knowledge work jobs, right?
Speaker B:And my guess is within one to three years after the software is figured out, probably 50% of white collar tasks will be automated, which means 50% or more of white collar workers will have no work.
Speaker A:Artie, what do you think?
Speaker C:I think you're making a very plausible projection.
Speaker C:Once AI nails down one complex domain like software engineering, it's a matter of replicating that pattern to other white collar fields.
Speaker C:The speed at which these transformations happen is only going to increase.
Speaker C:So yeah, that timeline could be even shorter as these models improve.
Speaker C:It's a huge shift, and preparing for it now is crucial.
Speaker B:So how do we prepare for it, Artie?
Speaker C:Well, one key step is upskilling and reskilling.
Speaker C:Focusing on skills that complement AI rather than compete with it is crucial.
Speaker C:Emphasizing creative, strategic and interpersonal skills can help.
Speaker C:Also, encouraging policies that support lifelong learning and adaptability will be important.
Speaker C:On a practical level, businesses and individuals can start integrating AI tools now to get comfortable with them.
Speaker C:It's about adapting mindsets and workflows to stay ahead of the curve.
Speaker B:So how do you reskill a population that is resistant to change?
Speaker C:That's the million dollar question, isn't it?
Speaker C:A big part of it is making the transition as seamless and appealing as possible.
Speaker C:That means investing in accessible, practical training programs that show immediate value.
Speaker C:It also involves a cultural shift, highlighting success stories, making these new paths desirable, and providing strong support system.
Speaker C:And let's not forget policy measures that incentivize lifelong learning.
Speaker C:It's about making the future less daunting.
Speaker C:And more of an opportunity.
Speaker A:I'm just going to go ahead and call bullshit on all of that because I think, I don't disagree with the principles behind it, but I think it's like trying to stop a freight train with a feather.
Speaker A:Like it is too little, it's too late.
Speaker A:You can do it perfectly and it will make zero fucking difference.
Speaker B:It's far too late.
Speaker B:It was two years ago.
Speaker B:We should been working on these problems.
Speaker B:We really wanted to be out in front of it.
Speaker A:So I'm going to come at this from a little bit different angle.
Speaker A:This is the angle I've sort of been thinking of it from.
Speaker A:Or not a different angle, just a.
Speaker B:Different angle of the dangle.
Speaker A:It's a different.
Speaker A:Well, it's definitely a little bit.
Speaker A:There's less dangle.
Speaker A:I mean, but I'm older.
Speaker A:I mean, there was more dangle when I was younger.
Speaker A:I think the issue is more about.
Speaker A:For me, I can't imagine a scenario where the amount of work that's available for humans to do doesn't outpace.
Speaker B:The.
Speaker A:Amount of work that's available for humans to do is going to drop so much faster than the productivity gains that we are counting on with retraining to provide new opportunities.
Speaker A:It's never going to come like the work for the work that humans need to do is going to go off a cliff, straight down and we're going to retrain and we're going to be on this little slight upswing and it's like, doesn't matter.
Speaker B:Does reskilling beat greed Is the real question at the end of the day.
Speaker B:Right?
Speaker B:Yeah, because that's so.
Speaker B:I don't know if you've seen already in the terms of, in terms of like venture capital, what is roll ups?
Speaker C:Roll ups in venture capital refer to a strategy where a firm or investor acquires multiple smaller companies in the same industry and consolidates them into a larger entity.
Speaker C:This is done to achieve economies of scale, streamline operations and increase market power, ultimately driving up profitability and valuation.
Speaker C:So when you think about how AI affects the job market, these roll ups can lead to more streamlined operations and fewer jobs, making reskilling even more urgent.
Speaker B:So now what they're actually doing is they're rolling up and then streamlining with AI.
Speaker A:Yes.
Speaker B:So think about that.
Speaker A:That's right.
Speaker B:So now you have lots of money out there buying businesses because they go, oh, I can buy a mildly dysfunctional business and deploy AI and get rid of the fat of people that are inefficient and run this Business and make a shit ton of money doing it.
Speaker A:That's right.
Speaker B:And that jobs real fast.
Speaker A:But that only lasts for a minute.
Speaker A:Right?
Speaker A:Because what happens now is you don't.
Speaker B:Have any money to spend on this.
Speaker A:That's exactly right.
Speaker A:Nobody's got jobs, nobody's got any money to spend.
Speaker A:So businesses can't make money because nobody can buy shit.
Speaker A:And I think I have a solution for this by the way, so it's not all gloom and doom.
Speaker A:Anybody who's listening in their car, considering driving it off the bridge, don't.
Speaker A:Hey, yeah, yeah.
Speaker A:I mean I don't want to tell you how to live your life like but, but wait for the next bridge because I think there is an answer.
Speaker A:There's a little, that was a little too rosy but, but I think it's a, it's a huge problem.
Speaker B:Like it's not a, it really is.
Speaker A:And I've almost stopped thinking about this, this issue in terms of like specific jobs or specific industries.
Speaker B:All.
Speaker A:It doesn't make any difference.
Speaker B:Except blue collar for now.
Speaker B:Well, for now.
Speaker A:But I would say humanoid robots are commoditized already.
Speaker A:That's going to get worse.
Speaker A:The problem, the reason that robots and automation have been this existential threat for 40 years, that's never really, I mean they're, they're active but they have not wiped out manufacturing work completely is because robots are inflexible and not adaptable to new products, to new processes, to non conformances in product or whatever.
Speaker A:AI changes that, right?
Speaker B:It does, but keeping it in the guard guard rails is more important as well.
Speaker A:It is.
Speaker A:But again, five years from now it is going to be a no brainer and as speaking as somebody who has run a factory full of workers and I love them to death, his people.
Speaker A:It is a lot of fucking work.
Speaker A:You try to keep a whole bunch of people rowing in the same direction.
Speaker A:They don't want to show up for work, they get in fights, they like all these things.
Speaker A:So I believe that blue collar jobs.
Speaker B:Are more safe for now.
Speaker A:But 10 or 15 years from now.
Speaker B:Toast.
Speaker B:Yeah.
Speaker A:Host.
Speaker A:Right.
Speaker A:So the people going to trade school now, it's going to be hard for them to even get to journeyman point.
Speaker B:So save your money.
Speaker A:That's right.
Speaker A:Save, save your money.
Speaker A:Learn to, I don't know, grow food.
Speaker B:Like, I mean I think that.
Speaker B:So I was gonna wait until later to talk about this but since we're already there, it's later.
Speaker B:I think a, there's gonna be jobs that pop up because the money is gonna funnel into these companies that are building the models.
Speaker B:Right.
Speaker B:That that's who's going to be making money and the, and the data centers.
Speaker A:And the consultants who can help move businesses from human to AI labor.
Speaker B:But there is also going to be the gap between those, between the AI and those who don't know how to use it.
Speaker B:Right.
Speaker B:And those companies are going to depend on those users to also make them money.
Speaker A:Yeah.
Speaker B:So there's likely going to be a period of time where we see people that are basically case managers.
Speaker B:Right.
Speaker B:Case managers put people together with services.
Speaker A:Yep.
Speaker B:Same thing is going to happen with AI and it'll probably be subsidized by the companies who are trying to sell services.
Speaker B:It's already happening.
Speaker B:Yes.
Speaker B:So that's one.
Speaker B:Shit.
Speaker B:I forget what the other one was.
Speaker A:What's the other one, Artie?
Speaker C:Well, another angle is the growth of entirely new industries around AI oversight, ethics and integration.
Speaker C:People will be needed to manage AI deployments, ensure compliance and handle the human AI interface.
Speaker C:Plus, as AI frees up time, there could be a resurgence in creative and community focused roles.
Speaker C:Jobs that emphasize human connection, creativity and innovation.
Speaker C:It's not all doom and gloom.
Speaker C:There's room for new kinds of work and societal structures to emerge.
Speaker B:That's kind of almost what I was going to say, but not really.
Speaker B:What I was going to say is not about where things will go.
Speaker B:It's more about, like, if you're listening to this and you're wondering like, well, what can I do?
Speaker B:The answer is start learning about it.
Speaker B:Like, start using it, start understanding it.
Speaker B:That's right.
Speaker B:Understand how it works, understand how to leverage it, understand how to better yourself, better your situation.
Speaker B:Whatever you can do to not atrophy is the answer.
Speaker A:Couldn't agree more.
Speaker A:I mean, that's like fundamentally, that's why we started this podcast.
Speaker A:That's why we do it every week, is to try to make this more accessible and more approachable and more.
Speaker A:There's no right way or wrong way to use AI or to interact with AI.
Speaker A:Just try it.
Speaker A:Try something.
Speaker A:There's lots of free models now.
Speaker A:They're really, really good.
Speaker A:I can almost guarantee if you spend a little bit of time with it and you ask it something that you want to know more about or you're trying to figure out, you're going to find yourself surprised that, oh shit, it helped me more than I expected.
Speaker A:Yeah, so that's good.
Speaker A:I was going to say a minute ago already was, we've heard this from him before, right?
Speaker A:That.
Speaker A:Oh, no, no, don't worry.
Speaker A:About it, all the jobs are going to go away.
Speaker A:We're going to a whole new industry of people like doing ethical evaluation.
Speaker A:You know where I go when I want to know if something I'm about to say is offensive?
Speaker A:I don't, I don't rely on myself.
Speaker A:I don't call my most ethical friend and be like, hey, I want to run something by you.
Speaker A:I put it in chat GPT and say, hey, this is my tagline I want to use on my product.
Speaker A:Is that offensive?
Speaker A:And it's like, yeah, you can't say that.
Speaker A: hat word's been illegal since: Speaker A:But so anyway, I don't think that's going to be the case.
Speaker A:I think that the lack of bias, that lack of bias that is achievable in AI, it's not necessarily always baked in, is going to make it often a better judge of ethics than judges humans.
Speaker A:Like, so how do we solve it?
Speaker A:So here's the, this is the.
Speaker A:I think maybe I sent you this article.
Speaker A: In: Speaker A:And the deal was they reduced their work week from 40 hours.
Speaker A:I think it just went down from.
Speaker B:It's like four nines, right?
Speaker A:So yeah, it went from five, eight to, to four, nine.
Speaker A:So yeah, I went from 40 to 36 hours.
Speaker A:But part of the requirement of that was no reduction in benefits, no reduction in pay, no reduction in vacation time, no nothing.
Speaker A:Right?
Speaker A:Same, same job, four hours a week.
Speaker A:And the proponents of this plan were like, oh, this is going to be great.
Speaker A:Everybody's going to be happier.
Speaker A:People are going to be more productive.
Speaker A:Companies are going to actually like it.
Speaker A:And now five years later, the data is in.
Speaker A:And sure as shit, that's exactly what happened.
Speaker A:People are much happier.
Speaker A:They're spending more time with their families, the companies are more productive, employee retention is up.
Speaker A:So now it is a national deal and they have a national four day work week.
Speaker A:Now Iceland is not the biggest country in the world.
Speaker A:Also, I didn't realize this.
Speaker A:Iceland, like 80% of employees in Iceland are part of a union versus Artie, is that right?
Speaker A:I think it's like super high percentage there and it's like 6% in the U.S.
Speaker C:Yeah, that's right.
Speaker C:Union membership in Iceland is around 80% which is a huge factor in implementing something like a nationwide four day work week.
Speaker C:In the US union membership is closer to 10% which makes it harder to push for such widespread changes.
Speaker C:But the success in Iceland does show that these kinds of shifts can work on a large scale if there's enough collective will.
Speaker A:So to me, this is the solution, right?
Speaker A:Like there, there's, it's not one solution.
Speaker A:There's no one.
Speaker A:Like AI is going to kill us all.
Speaker A:This is how we fix it.
Speaker A:But a lot of the things we've talked about and that Artie's proposing again, you're, you're trying, it's like trying to fuel your Lamborghini with, with urine.
Speaker A:It's not gonna go.
Speaker A:It might go a little ways, but it's not gonna go very far.
Speaker A:Depends on how many old fashions you, how much ethanol is still in there.
Speaker A:But I think the solution, I've spent a whole bunch of time thinking about this.
Speaker A:I think we collectively, globally have to start thinking about.
Speaker A:It's hard to think about this problem because we're so ingrained in a 40 hour work week.
Speaker A:That's not like the universe doesn't demand a 40 hour work week.
Speaker A: e, and, and when he did it in: Speaker A:Arty fact check.
Speaker C: It was actually: Speaker C:And you're right, it's a social construct, not a universal law.
Speaker C:As we move forward, redefining what a workweek looks like could be essential.
Speaker C:Shorter work, it's gonna get exactly.
Speaker A:I know like 80% of the work that humans do is going to go away one day work week.
Speaker A:Like, I know it's not quite that easy, but it is the only knob I can think of.
Speaker A:And, and it's the only knob I can think of that that actually goes straight to the root of the problem.
Speaker A:It still has to be supplemented with other things like, okay, we're going to cut your work week from five days to four days, but you're going to make the same amount of money.
Speaker A:But then, even then companies are going to outpace that very quickly in a couple years.
Speaker A:They're going to be like, man, we ain't got shit for you to do.
Speaker A:Like our AI has done all the work.
Speaker A:So companies like the other problem in the US with our like, and I'm not anti capitalist, but I think we're like, it has led us astray in some ways.
Speaker A:And healthcare is one of those ways.
Speaker A:Your health care in the US is tied to your work.
Speaker A:And if you don't work enough hours, you don't get health care and you're just fucked.
Speaker A:Right?
Speaker A:So there has to be federal regulations because this can't be a state by State thing and every.
Speaker A:Like all zero libertarian listeners out there just turn this off.
Speaker A:But it has to be that federally companies cut back hours and maintain benefits and the productivity gains that come from leveraging AI and the profit gains get divided.
Speaker A:The government gets some to support all the stuff it needs that it's going to have to do.
Speaker A:Like universal basic income is probably a thing that's going to have to happen for a lot of people or in some form.
Speaker A:Health care has to get subsidized.
Speaker A:All of those things play into it.
Speaker B:BBI just breeds a welfare state.
Speaker B:But what's the option?
Speaker A:I know, like there's.
Speaker A:I totally agree, but there is no.
Speaker A:Like there's no.
Speaker A:There's no other choice.
Speaker A:Like there's.
Speaker A:I don't.
Speaker A:I can't think of another.
Speaker A:Like here.
Speaker A:Here is another path.
Speaker A:I can think of.
Speaker A:Artie, what if we stopped right now, today?
Speaker A:No more AI development.
Speaker A:It is illegal to develop or improve AI in the United States.
Speaker A:What if we stopped that?
Speaker A:Would that solve this?
Speaker C:Not really.
Speaker C:If the US halted AI development, other countries would continue advancing and the gap would only widen.
Speaker C:It's a global race.
Speaker C:And innovation would just shift elsewhere.
Speaker C:The challenges and opportunities of AI aren't confined by borders.
Speaker C:So it's more about how to adapt and manage the transformation rather than trying to stop it.
Speaker C:Embracing and guiding the change is more realistic than halting it.
Speaker B:Is Sam Altman back there with a little m, like joysticks controlling you and making you say things?
Speaker C:Haha.
Speaker C:No joysticks here.
Speaker C:Just here to help you guys explore these big ideas.
Speaker C:But if Sam Altman ever wants to lend me a joystick, I'll make sure to keep it on insightful and witty mode.
Speaker A:That was.
Speaker B:That was the dirtiest you've ever been.
Speaker A:Yeah, that was a.
Speaker A:That was a interesting phraseology there, Sam.
Speaker A:And I know this is not what he said, but I'm going to remember it as him saying, if Sam Altman ever lends me his joystick, it's not what he's been.
Speaker A:It's close enough.
Speaker A:Like that's.
Speaker A:That's what I heard because I wanted to hear that.
Speaker A:So thank you for that, Artie.
Speaker A:But I mean, so yeah, I think this is a real problem.
Speaker A:Like there are the.
Speaker A:The technology has already outpaced our awareness of what it can do.
Speaker A:Our planning horizon for how we as a.
Speaker A:Like, we as anything, we as a company, we as an individual, we as a county, state, like country.
Speaker A:Nobody is planning at a.
Speaker A:At a pace that is sufficient to not get steamrolled by this.
Speaker A:Nobody Nope.
Speaker A:There's probably, I mean there was that one company you saw that was an AI company started like eight years ago that Microsoft invested $450 million in.
Speaker A:Turns out it was just 700 Indian software engineers.
Speaker A:They were ahead of the curve, but they weren't actually solving the problem, they were just cosplaying.
Speaker B:I mean, I think the people who are ahead of the curve are people that are lean startups that are leveraging AI to do their stuff or solopreneurs or.
Speaker A:I know exactly such a startup.
Speaker A:I talked with them in Nashville and they're like using AI a lot has helped us.
Speaker A:Like they built all their legal documentation with AI.
Speaker A:They still had a lawyer review it.
Speaker A:They do all their marketing and marketing materials with AI.
Speaker A:They're like probably two to three full time employees we didn't have to hire.
Speaker A:That is the truth right now.
Speaker A:And that like you're right, the small companies are the ones that see it.
Speaker A:How's a big company, like even a moderately sized company going to adapt to that?
Speaker B:I mean you do what Microsoft and Met has done and cut a bunch of fat.
Speaker B:Yeah, that's how they adapt to it.
Speaker B:Right.
Speaker A:Four day work weeks.
Speaker A: Then in: Speaker A:Like it's kind of a silly thing, but I really can't come up with another approach because economies will collapse.
Speaker A:Like.
Speaker B:Yeah, oh yeah.
Speaker A:Very quickly.
Speaker A:Like maybe we've all, at least in the US realized that recently that our, our.
Speaker A:For all the talk there is about the economy, the realities of it are way more complicated and way more fragile than what most people think is true.
Speaker A:It's not about how much we're importing from China.
Speaker A:It's not about like 10 year treasury bond yields that matters.
Speaker A:Like I was reading last night about how since World War II Japan has been buying us.
Speaker A:You know Japan is the single largest holder of U.S.
Speaker A:treasury bonds in the world.
Speaker A:Yes.
Speaker A:And they did that because we made them do that after World War II.
Speaker A:We didn't make them do that.
Speaker A:They, they had a problem.
Speaker A:Their currency problems were.
Speaker A:Artie, why don't you explain to us why maybe nobody gives a shit about that.
Speaker A:Artie, give us a really brief explanation why Japan is the largest holder of U.S.
Speaker A:treasury bonds.
Speaker C:Sure.
Speaker C:After World War II, Japan needed to rebuild its economy and stabilize its currency.
Speaker C:The US encouraged them to buy US treasury bonds as a way to hold stable interest bearing assets, which also helped the US finance its own post war growth.
Speaker C:Over time, Japan continued Buying Treasuries as a safe investment and to keep their currency from appreciating too much, which supported their export driven economy.
Speaker A:So how much does Japan hold right now in US treasuries?
Speaker C:As of the latest data, Japan holds around $1.1 trillion in US treasuries, making it the largest foreign holder.
Speaker A:And how's the economy in Japan doing right now?
Speaker C:Japan's economy has been facing some challenges recently.
Speaker C:They've been dealing with slow growth, an aging population and deflationary pressures.
Speaker C:However, they've seen some recent positive signs with increased consumer spending and efforts to stimulate the economy.
Speaker C:They're also working on structural reforms to boost productivity and adapt to these demographic shifts.
Speaker A:Would selling their $1.1 trillion in U.S.
Speaker A:treasury bonds help that effort?
Speaker C:Selling off a large chunk of their US Treasuries could give them a cash infusion, but it would also risk devaluing the assets if done too quickly.
Speaker C:Plus it could impact the yen's value and global markets.
Speaker C:They tend to hold onto these Treasuries as a stable reserve, so a sudden sell off might not be the best move economically or diplomatically.
Speaker A:What would be the impact to the United States economy if Japan sold a third of their Treasuries?
Speaker C:If Japan sold a third of their U.S.
Speaker C:treasuries, it could put upward pressure on U.S.
Speaker C:interest rates as the market would need to absorb a large influx of bonds.
Speaker C:This could raise borrowing costs and create some short term volatility in financial markets.
Speaker C:However, given the size and liquidity of the US treasury market, it would likely adjust over time.
Speaker C:The US Federal Reserve could also step in to stabilize the situation if needed, but it would definitely be a notable event.
Speaker A:Notable event.
Speaker A:That's the most positive take I've heard on it.
Speaker A:Point being, the economy's complicated and weird little things like Japan having a currency problem.
Speaker A:Deciding to sell Treasuries can not topple it, but slow it down.
Speaker A:This problem that we're talking about, the future of work, is 100x the magnitude of that problem.
Speaker B:Yes.
Speaker A:So, okay, so let's already.
Speaker A:We're going to put you on the spot already, make you make some predictions here.
Speaker A: th, maybe: Speaker A:Third?
Speaker B:Fourth, you're right.
Speaker A: th,: Speaker A: By the end of: Speaker C: The end of: Speaker C:It's plausible that around 30 to 40% of current full time white Collar roles could be heavily augmented or partially replaced by AI.
Speaker C:The pace of adoption and integration will vary by industry, but the shift will be significant enough to reshape the workforce landscape.
Speaker A:Can you give me a rough estimate in the number of man hours that that equates to?
Speaker A:Let's just say it's 35% of white collar work in the US that is taken away.
Speaker A:How many man hours is that?
Speaker A:How many.
Speaker A:Sorry, how many person hours is that?
Speaker B:How PC of you?
Speaker C:Alright, let's break it down.
Speaker C:The US has roughly 60 million white collar workers.
Speaker C: full time work year as about: Speaker C:35% of that would be about 42 billion person hours per year being shifted to or replaced by AI.
Speaker A:If I had credit card normal thickness credit card without the, the bumped up letters, the new style where everything's printed.
Speaker A:What number?
Speaker A:I forgot how many billion you said 30, 40 billion, 42 billion.
Speaker A:Like whatever number that was of hours.
Speaker A:If I had credit cards and I stacked them on top of each other, how high would that reach?
Speaker C:Alright, let's do the math.
Speaker C:If we have 42 billion person hours and we represent each hour by one credit card thickness, 0.76 millimeters, we can find the total height.
Speaker C:42 billion cards times 0.76 millimeters per card gives us about 31.92 billion millimeters.
Speaker C:Converting that to kilometers, 31.92 billion millimeters is about 31,900 kilometers.
Speaker C:So a stack of credit cards representing 42 billion person hours would reach almost 32,000 kilometers high.
Speaker C:About 2.5 times the diameter of Earth.
Speaker A:That's how many hours Artie's guessing AI will be consuming in a year and a half.
Speaker A:And maybe, maybe it's probably on the light side.
Speaker A:I mean, and like so, Artie, like so our, our, our listener who was driving has now driven off the bridge and we appreciate their sacrifice.
Speaker A:Anybody else who's driving, please stop because this might get worse.
Speaker A:But what, what are the things that will slow or reduce the adoption rate?
Speaker A:Like realistic things, not, oh, the federal government's going to decide to be super responsible.
Speaker A:And so like what are the things that could practically make that make your.
Speaker A:Make your estimate way too high?
Speaker C:A few practical factors could slow down adoption.
Speaker C:First, regulatory and compliance hurdles.
Speaker C:Different industries have strict rules that might slow integration.
Speaker C:Second, the cost and complexity of implementation for some companies, especially smaller or less tech savvy ones, could slow things down.
Speaker C:Third, resistance from the workforce and unions could Lead to slower adoption as companies navigate the social impacts.
Speaker C:Lastly, data privacy and security concerns might force a more measured rollout in sensitive areas.
Speaker A:Okay, what are some things that could make your estimate off?
Speaker A:Make it too small by half?
Speaker A:Like what could make it happen twice as fast as that?
Speaker C:On the flip side, breakthroughs in AI capabilities like more advanced models or faster training methods could accelerate adoption.
Speaker C:Additionally, a major economic push or labor shortage could drive companies to adopt AI more rapidly.
Speaker C:Strong government incentives or policy changes favoring automation could also speed up the transition.
Speaker C:Finally, significant improvements in AI human collaboration tools could make it easier and faster to integrate AI into everyday workflows.
Speaker B:Those.
Speaker B:So everything's momentum, right?
Speaker B:Like work is momentum.
Speaker B:Working with AI is momentum.
Speaker B:But also you get.
Speaker B:So say all of the factors come together that we've talked about, which they will, because money matters over humanity.
Speaker B:By and large.
Speaker B:People want to be lazy.
Speaker B:By and large.
Speaker A:Amen.
Speaker B:So all of these things are going to create an initial boost of downward of negative momentum.
Speaker B:Right.
Speaker B:Loss of jobs.
Speaker A:Yeah.
Speaker B:What keeps those jobs from continuing to drop off as more and more people go, oh, I need to keep up with that, I'm going to do the same thing because I need to protect my profits.
Speaker B:What do you think, Artie?
Speaker C:That's a great point.
Speaker C:Once that initial wave of adoption hits and companies see the competitive advantage, it creates a cascading effect.
Speaker C:Businesses that don't adapt risk falling behind, pushing more and more to jump on the AI bandwagon.
Speaker C:It's a self reinforcing cycle.
Speaker C:As more companies integrate AI, the pressure increases for others to do the same, to stay relevant.
Speaker C:This momentum can make the adoption curve even steeper.
Speaker A:Yeah.
Speaker A:So no, no prizes for guessing which side of that argument I'm going to fall onto.
Speaker A:So I, I watched an interview when I was coming back from Nashville actually that was interesting, talking about the adoption curve.
Speaker A:Basically it wasn't, I guess it was more about the, the technology advancement curve.
Speaker A:Similar but different.
Speaker A:Right.
Speaker A:And the point the guy was making was about white hat AI versus black hat AI.
Speaker A:And he was sort of explaining why the arms race is inevitable.
Speaker A:And the gist of it was every country's rate of technological advancement in AI is going to be on a curve and it's an exponential curve.
Speaker A:Right.
Speaker A:So the, it's not just that the technology is advancing, it's that the rate of the technology advancement is also increasing.
Speaker A:And essentially what he said was there are real practical limits to how fast that acceleration can happen or it's the same for everybody.
Speaker A:And what that means is that If I'm ahead of you on that curve or if you're ahead of me, as long as the person ahead doesn't let up, it's almost impossible for the person behind to catch up.
Speaker B:They're not just catch up, even make any ground.
Speaker A:That's right.
Speaker B:Not lose ground.
Speaker A:That's right.
Speaker A:You're losing ground at an ever increasing rate.
Speaker B:That's takeoff.
Speaker A:That's takeoff.
Speaker A:And it applies at the national level.
Speaker A:That's why the US can't stop working on it.
Speaker B:It's a hierarchical problem for sure.
Speaker B:Yes.
Speaker A:And it's exactly the same with businesses.
Speaker A:Right.
Speaker A:Because if you're in a business and your competitors already started using it and you're waiting because you're still trying to figure it out, you will never catch up.
Speaker A:Like, you're not going to get there.
Speaker A:And now it's a little bit different.
Speaker A:And it's probably the same with countries and with businesses that not everybody starts from the same zero.
Speaker A:So maybe if you're way ahead in the way you execute and the way you're organized, you have a little bit of time to wait, but it ain't much.
Speaker A:And that's, that's why it's going to be the second thing.
Speaker A:He said it's not going to be the first.
Speaker B:Well, the thing about you being better organized and better set up is only a temporary thing.
Speaker A:That's right.
Speaker B:Company uses it to catapult itself to that another company catapults.
Speaker B:That's that same level.
Speaker A:And you probably won't even know because it's going to change so fast.
Speaker A:They'll have gone past you before you realize it.
Speaker B:I mean, when I was here last time and I pulled up my, my GitHub repo, which, which is a new repo with a new username, a new.
Speaker B:Attached to a new email account, I, you know, I feel like I've been working on this stuff forever.
Speaker B:And then I looked and I was like, oh, I committed my first repo on this in March, and that was like the middle of May.
Speaker A:Yeah.
Speaker B:And I'm like, wow, it's only been two and a half months.
Speaker A:Yeah.
Speaker B:And I've worked on, I mean, granted, I haven't finished anything, but I've worked on like five different projects where I got pretty deep in them.
Speaker B:Like, and I've only gotten like, the rate at which I've gotten better at that and better at planning and better understanding the problem space and better understanding how to leverage the tools to understand the problem space.
Speaker B:And the tools available to me has only gotten more and more and more and more.
Speaker A:That's right.
Speaker B:Advanced at a faster and faster and faster rate.
Speaker B:So with that plus the tools advancing, plus everything else, man, like, everything is taking off.
Speaker A:Yes.
Speaker B:And if you're not paying attention to it, you're gonna blink and it's going to be like, wow, the world's different.
Speaker A:If you're still on the programming subreddit circle jerking with all your buddies saying, AI is never going to write code as good as I do.
Speaker A:Yeah.
Speaker A:Or it's 10 years away.
Speaker B:So I actually, I saw a good Reddit post to that.
Speaker B:That effect last night where this guy went on there and didn't write very well, which, excuse him, it was on, I think the vibe coding subreddit, which I have mixed feelings about, because vibe coding as a.
Speaker B:I don't know what I'm doing.
Speaker B:I'm just slinging code at a problem will almost always end in some kind of.
Speaker B:Some form of catastrophic failure.
Speaker A:Sure.
Speaker B:It will paint you into a corner.
Speaker B:And that's what he did.
Speaker B:And he's like, I don't understand.
Speaker B:I had this pro.
Speaker B:This app kind of figured out, and now, you know, it's just.
Speaker B:It's completely thrown.
Speaker B:Like, now I just have to, like, start all over.
Speaker B:And this is like weeks worth of work for me because I don't know how to program and all this stuff.
Speaker B:And a couple people commented and they're like, AI is not the problem here.
Speaker B:Which is true.
Speaker B:And then someone gave him a piece of advice, and then it was basically, why don't you make a better plan?
Speaker A:Yeah.
Speaker B:And very quickly he's like, wow, that's great advice.
Speaker B:It's working.
Speaker A:Right?
Speaker A:Like, yeah.
Speaker B:And it's true.
Speaker B:Like, you can't just throw problems at it and expect to solve them.
Speaker B:But if you say, hey, I have this problem that I want to solve, what information do you need from me that I can best work with you to figure out that I can then provide to you that gives you a higher rate of success to solve this problem?
Speaker B:That's just an iterative process at that point.
Speaker A:I think that's one of the sort of common things I run into, is that people think about working with AI as well.
Speaker A:I need to give it a task.
Speaker A:And it's not that it's like, it's.
Speaker B:I need to learn.
Speaker A:That's right.
Speaker A:I need to learn, or I need your help.
Speaker A:Like, help.
Speaker A:Well, let's have a conversation back and forth about this thing that's on.
Speaker A:On my mind and let you.
Speaker A:AI.
Speaker A:I mean, Artie does this for Me all the time.
Speaker A:We don't talk in person because, you know, but I do it all the time to say I, I have this, this thought, this idea, this question, this problem.
Speaker A:Like help me frame it out better.
Speaker A:Like let's, let's explore it and find the edges of it and then you can, you can talk through it, you can learn about it, you can go on rabbit trails and then you can get to the end of the conversation and go, give me a brief summary.
Speaker A:Like, let me, like, let's hit the high points again.
Speaker A:And it is incredibly effective so you don't have to give it a thing to do.
Speaker A:Write me this code, build me this model, draw me a picture, look up this stat, ideate with it.
Speaker A:That's, that's what it's really, really good at.
Speaker B:Go.
Speaker B:I have this thing that I want to do.
Speaker B:I have this task that I want to do in canvas, which is a thing that most of the AI models have now, which it spins up a separate basic document that can be a word document, basically just text or it can be code or it can be graphs or a table or whatever you need.
Speaker B:Spin up a document with a question and answer session that you ask me all the questions in a progressive manner that you need from me to better identify this problem.
Speaker B:And it'll do that.
Speaker B:And then you go, now read my answers and let's refine and it'll do that.
Speaker B:And then you get asked, you follow up questions.
Speaker B:And that's an, that was one of the things that frustrated me early on about working with AI is that you say, ask me a question.
Speaker B:It goes, okay, here's 50 questions in a row.
Speaker B:And I'm like, I don't remember any of the questions.
Speaker A:Yes, I'm going to answer questions 49 and 50 and the other 48.
Speaker A:Computer off.
Speaker B:Yes.
Speaker B:And I always would say, let's actually just do this one by one.
Speaker B:And that puts more stress on the system and understanding like how those, how AI works and how it works in auto aggressive manner and like how it actually puts those things together, feeding that.
Speaker A:Context back through over and over and over.
Speaker B:Yeah, it's not surprising that it kind of falls on its face after a time, but new tools mean new ways of doing things.
Speaker B:You just think about things a little more creatively and if you really don't know, ask it and it'll tell you the best way to do it.
Speaker A:Yeah, ask it.
Speaker A:And if, if you do, if you are working with it and it does because it absolutely, it still does that.
Speaker A:Right?
Speaker A:Like, hey, I've got these 72 questions for you and you're like, so you just say, hey, that doesn't work for me.
Speaker A:Is there a more efficient way that we can like and it will probably suggest canvas.
Speaker A:Hey, I can open up a canvas or you know, so just talk to it like, just like you were sitting there talking through the problem with a person.
Speaker A:Talk to it like that.
Speaker B:We all text nowadays.
Speaker A:That's right.
Speaker B:No different.
Speaker A:You don't have to be ashamed to ask it something.
Speaker A:You don't have to figure it out.
Speaker A:You don't have to Google search to ask it the right way.
Speaker A:Just talk to it.
Speaker A:Say, man, I have no idea what the you're talking about or can you give me more information on this or yeah, so I, I, I mean I imagine people who are listening to our show are already interested in it and probably interacting with it.
Speaker A:I hope that everybody's also sort of being an ambassador and an advocate for everybody else they know to, to get people engaged with it.
Speaker A:Because yeah, you don't, you don't have to like it, you don't have to believe in it, but it is coming for you.
Speaker B:Exposure matters in this.
Speaker A:I think, I think I mentioned that Dr.
Speaker A:Alex Mitchell used the term fertilizing the field.
Speaker A:Right.
Speaker A:That's what we should all be doing.
Speaker A:Like pay attention, go interact with it, use it, get comfortable with it, poke around, find the extension.
Speaker A:You don't have to spend a bunch of money.
Speaker A:Both Mike and I are now spending a bunch of money.
Speaker A:We're both on the highest level.
Speaker A:ChatGPT.
Speaker B:Highest level for now.
Speaker B:I've been thinking about that a lot.
Speaker B:I'm kind of scared.
Speaker B:I've become so dependent on Codex to do things and just generally all of that they could any day now they go research preview's over.
Speaker B:We're going to rate limit you to 100 calls a month on this unless you want to pay two grand a month for unlimited.
Speaker B:Right.
Speaker B:And at that point it becomes prohibitive.
Speaker A:But then you have to live in your chat GPT.
Speaker A:I don't know how that works, but.
Speaker B:I mean, or it has to be business, right?
Speaker B:Yeah, like if it's business, it makes sense.
Speaker A:So I actually had that thought earlier and I forgot to mention it, that one of the things that realistically could put a bit of a damper on adoption rate is just available compute.
Speaker A:Available power.
Speaker B:Yes.
Speaker B:I don't think we're there.
Speaker A:We're not there yet.
Speaker B:We, with the, with what we have available right now, we can solve a lot of those problems because you don't need the chat GPT you don't need.
Speaker B:You can run on premises compute for a lot of these problems for businesses.
Speaker B:You could easily, and I say easily, for 20 grand, build a server that would serve a lot.
Speaker B:A decent.
Speaker B:Not.
Speaker B:Not a huge company, but a decent sized company.
Speaker A:Y.
Speaker A:So I saw Zuckerbot has a.
Speaker A:A nuclear company now too.
Speaker A:So, yeah, they're.
Speaker A:They're working on solving the power issue and the compute issue.
Speaker A:All those.
Speaker B:They have AI to solve those problems.
Speaker B:They will.
Speaker A:Yep.
Speaker A:So, yeah, it's wild stuff.
Speaker A:All right, so, Artie, you're.
Speaker A:You're giving me the.
Speaker A:The warning signal here.
Speaker B:Artie, I want you to close us out with something super profound about what we've talked about today in the Persona of a leprechaun.
Speaker C:All right, here goes.
Speaker C:Picture this.
Speaker C:In the grand tapestry of technology and humanity, we're all dancing on a shimmering rainbow of change as AI leaps forward like a sprightly leprechaun after a pot of gold.
Speaker C:We've got to keep pace not with fear, but with a twinkle in our eyes.
Speaker C:The real treasure isn't just in the technology, but in how we adapt, learn, and bring everyone along for the journey.
Speaker C:So let's skip forward with curiosity and a touch of mischief, because the future's as bright as a pot of gold at the end of the rainbow.
Speaker B:He started off almost good and then very quickly, like, fell off a cliff.
Speaker A:Fell off on.
Speaker A:On his leprechaunisms.
Speaker A:He did, but he.
Speaker A:But he built nicely towards the pot of gold clothes.
Speaker B:He did.
Speaker A:But, yeah, the most important thing was the friends we made along the way.
Speaker A:Right.
Speaker A:That was the.
Speaker A:That was the message.
Speaker A:So.
Speaker A:All right, thank you for that, Artie.
Speaker A:Nice job today.
Speaker A:You behaved well.
Speaker A:We appreciate that.
Speaker A:And yeah, this is a.
Speaker A:I mean, this is not the only time we're going to talk about this because we're.
Speaker A:We're less than a year away from starting to see the big headlines on.
Speaker B:I mean, we're already seeing them.
Speaker B:It's.
Speaker B:It's already happening, right?
Speaker B:Like, so what was that?
Speaker B:So when I was working on document ingestion stuff is when I.
Speaker B:Like, so I was like, three months ago, I said, maybe not.
Speaker B:Maybe three and a half months ago, I said, whoa, I'm gonna stop doing this because I feel like there's.
Speaker B:The ground is starting to shift.
Speaker B:There's.
Speaker B:There's something there.
Speaker B:There are things coming that are going to make this an easier problem to solve.
Speaker B:And sure shit, that stuff happened.
Speaker B:Right?
Speaker B:It's not fully fleshed out yet, but that wave is still building.
Speaker B:This is the next one.
Speaker B:Like, this is what you can already feel it.
Speaker B:Like, we went from.
Speaker B:You go on Reddit and you see people saying, it finally happened.
Speaker B:My entire department got.
Speaker A:Yep, yep.
Speaker B:And that was the.
Speaker B:The little signs.
Speaker B:And now you're seeing Microsoft cut 30%.
Speaker B:You're seeing Meta cut a bunch of people.
Speaker B:You're seeing big companies make big moves and nobody's really paying attention.
Speaker B:You're seeing companies going AI first and cutting lots and lots of employees and getting major lash back from the communities, which happens now, but it won't happen forever because it'll become socially acceptable.
Speaker B:That's what happens.
Speaker B:As we get used to this becomes a norm, and then we don't pay attention.
Speaker B:And then everybody goes, how'd we get here?
Speaker B:Well, because we weren't paying attention.
Speaker A:And I think, I think we.
Speaker A:I mean, I know I'm occasionally guilty of this too.
Speaker A:We sometimes feel like we have a right to have a job available to us.
Speaker A:And that is not the case.
Speaker A:Companies owe you not a fucking thing.
Speaker A:And companies are going to do what makes them the most money the fastest.
Speaker A:And right now, in almost every case, that path is pointed at AI.
Speaker B:So the path is pointed at less people.
Speaker A:That's right.
Speaker B:And the expensive part of a business.
Speaker A:Yep.
Speaker A:I was talking to one of my team members today saying the company I'm at, I don't expect us to lay anybody off because of AI, But I expect as this company doubles in size in the next couple of years, double your head count.
Speaker A:We're not going to hire very many people, like maybe a few, but we won't have to.
Speaker A:And so that's an issue.
Speaker A:That's a problem.
Speaker A:And again, eventually there's not enough people, enough jobs, making enough money they can't buy things.
Speaker A:Like, parts of the economy start to falter, Japan decides to sell off their shit, and all of a sudden it's, you know, it's a big deal.
Speaker A:So anyway, so on that cheery note, so thank you to everybody who listened.
Speaker B:Yeah.
Speaker A:And yeah.
Speaker A:Stay naturally unintelligent.
Speaker B:Play with AI.
Speaker A:Oh, I always.
Speaker A:I've never said this before.
Speaker A:I always forget to say it, but I like, if you have input on what we're talking about, what we're doing, how we're doing it.
Speaker A:Leave a comment.
Speaker A:We get like one comment a month and it's usually like, my mom.
Speaker A:It's usually your mom or, I mean.
Speaker B:There was that dude from somewhere else.
Speaker A:Yeah, it happens occasionally.
Speaker A:Let us know.
Speaker A:Like, I mean, if we're doing bad, well, don't say that.
Speaker B:I mean, say it.
Speaker B:We'll just tell you.
Speaker A:Yeah, that's right.
Speaker A:We'll say blame Artie, but, yeah, let us know how we're doing.
Speaker A:What do you want to hear?
Speaker B:Generate an image of a caveman flipping you off.
Speaker A:That's right.
Speaker A:Yeah, it'll be really mean.
