Cream City Calculation

Will AI take your job?

• Cream City Calculations

Share episode

There's been a lot in the news lately regarding AI replacing people's jobs. Frankie, Sal and Colleen break it down for you and provide their thoughts on the reality of this concept, and suggest ways to make yourself ready for a more AI-embedded world.

Articles referenced in this episode include:

These Jobs Will Fall First As AI Takes Over The Workplace
https://www.forbes.com/sites/jackkelly/2025/04/25/the-jobs-that-will-fall-first-as-ai-takes-over-the-workplace/

AI is ‘breaking’ entry-level jobs that Gen Z workers need to launch careers, LinkedIn exec warns
https://fortune.com/2025/05/25/ai-entry-level-jobs-gen-z-careers-young-workers-linkedin/

AI Won’t Take Your Job if You Know About IA
https://www.gse.harvard.edu/ideas/news/24/02/ai-wont-take-your-job-if-you-know-about-ia

AI won’t take your job; it will make you better at it
https://news.temple.edu/news/2025-01-16/ai-won-t-take-your-job-it-will-make-you-better-it

Behind the Curtain: A white-collar bloodbath
https://www.axios.com/2025/05/28/ai-jobs-white-collar-unemployment-anthropic

How AI is reshaping the career ladder, and other trends in jobs and skills on Labour Day
https://www.weforum.org/stories/2025/04/ai-jobs-international-workers-day/

Amazon boss tells staff AI means their jobs are at risk in coming years
https://www.theguardian.com/technology/2025/jun/18/amazon-boss-tells-staff-ai-means-their-jobs-are-at-risk-in-coming-years

Not an article, but just for fun - this has been around a long time, but “Will Robots Take My Job?”
https://willrobotstakemyjob.com/



Colleen:

Welcome to the Cream City Calculations podcast. We're three colleagues and friends that love data and to talk about how data is impacting our lives. I'm Colleen. I'm Frankie. And I'm Sal.

Sal:

Welcome to Cream City Calculations podcast. Today is gonna be a fun, kind of also dreary, outlook on, AI and how AI is, has a possibility to take our jobs. Is it gonna actually take care of jobs? What types of jobs does it take? We're gonna just discuss a couple articles, and then, kind of just take the insights of all three of us.

FRANKIE:

when I'm thinking about AI lately, like thinking about when we were kids and we were reading our history books and reading about the industrial revolution and some of that. And I'm thinking like, wow, like we're in the brain of like next industrial revolution. In my opinion. I think one day kids are gonna be reading about AI and what it did for all the different industries and thinking about it like as, whoa, that's crazy that that could ever happen. Right?

Colleen:

Yeah. Or, or that it's crazy that ever we have these jobs where we had to manually write code. I do think of things in terms of this is gonna be a dramatic shift and it's going to change everything

Sal:

yeah, and I think we're in this beginning stages even. Like what? Like reading the jungle or you know, industrial revolution back then. Looking at how that transition was rough

Colleen:

Yeah.

Sal:

of people. I think we're gonna go through that a little bit is how are we, like new jobs will be created, but people haven't created those jobs yet and people's jobs will be taken a little bit. And so we're gonna be in this transition

Colleen:

Yeah.

Sal:

To kind of evolve.

Colleen:

I actually think it's gonna be

Sal:

I,

Colleen:

I think of, what I remember from history class and the industrial revolution is maybe not as sharp as what you guys remember, but I remember talking about things like the invention of Henry Ford and how he automated his, production lines.

FRANKIE:

Okay.

Sal:

the line.

Colleen:

they assembled cars. I mean, that was a new concept, but still the same workers, could be trained relatively quickly

Sal:

be

Colleen:

on how to operate as part of an assembly line versus an individual person assembling an entire car themselves or however they did it before. Then I think this is much bigger and that there are going to be people who do not have these skills to keep up, that are gonna find themselves displaced and they will be able to find work again, if and when they sort of go through some retraining to learn AI skills, I think it's gonna be a bit more disruptive to anybody who thinks that it's not going to impact their jobs and thinks it's a tech subject, that they don't need to be techie, they don't need to know ai. I think it's gonna be more impactful to people like that.

FRANKIE:

I think there's probably some industries though that maybe won't be as affected.

Colleen:

Yeah,

FRANKIE:

the trades, like, you're

Colleen:

the trades.

FRANKIE:

still gonna need a plumber to come to your house or an electrician,

Colleen:

Yep.

FRANKIE:

things like that.

Colleen:

Yep. Yeah, healthcare. I think some of the articles I read, in preparation for this chat today, talked about healthcare and, like social work and things that are harder to automate. You know, those things won't be as affected. I think they'll still be elements of AI that you can work into them, which might be helpful to those people. But you're absolutely right. There's some industry is more impacted than others.

FRANKIE:

Yeah, I wonder what they'll call it,

Sal:

so

FRANKIE:

what are.

Sal:

A friend of mine, he is a doctor and he had a radiologist come and the AI system that they're using is using a, like it identified pneumonia or something, really small, but the doctor couldn't see it. And he's like, which do I. Diagnosed, but I don't see it on the, the x-ray. But do I diagnose that it's this'cause the AI's telling me it, there's just like this, like crazy like, oh, who do

FRANKIE:

Yeah.

Sal:

But yeah, like I do think it's actually gonna go into healthcare. I think there's gonna bes significant jobs that are enhanced by it. I

Colleen:

Right.

Sal:

'cause I don't know if fully if jobs will be replaced if people take it on.

FRANKIE:

I also heard the stories of, like at Mayo Clinic, how they've been able to diagnose things like cancer quickly or more quickly than what they do today. Just given, some of those AI images and being able to. An AI model, go through image after image. Human knowledge is literally like machine learning, right? Using the human knowledge, was it cancer or was it not cancer? And they just train it basically completely off of what the doctors had known in the past and what it was. I think that's really interesting. then these models are able to maybe catch something a little quicker than the human eye. Just kind of like what you were saying, Sal, with the pneumonia and maybe the doctor couldn't see it yet, but maybe the next day if they would've done an x-ray, it would've been more prevalent.

Sal:

Like one of the conversations that we had, and I know this is going a little deep, but it was about like how if something like malpractice, so if AI connect or finds something and the doctor doesn't believe it and doesn't react on that. Like, who is liable

Colleen:

Oh yeah, that's a good point.

Sal:

And so we were having that discussion, but it's like you still gotta trust the doctor and the human eye, to double check and maybe do additional testing. Maybe it forces additional testing, but you can see like the catalyst of that happens. Like, oh, this person now has to go and go through additional testing, more costs. It's like all because something. Spotted it earlier, which is great, but if it's wrong, you can see that there's a massive,

Colleen:

Yeah. Massive. I think risk on the part of that healthcare provider, whereas in the past, maybe, it was their discretion and if they really weren't sure, they could just wait to see if things got worse. But now if there's something to prove that, some AI model or something thought that this person had whatever disease earlier than what they got treated for it, you're right. I think that kind of opens things up. Like there's now proof saying, Hey, there was something to suggest that maybe we should have tested further for this. Earlier than what we did.

FRANKIE:

One thing about too, like how insurance companies might utilize AI as well and be like, you don't need to be treated for that.

Sal:

Exactly like episodic of care,

Colleen:

Mm-hmm.

Sal:

It's like the care plan. Like if a doctor doesn't do exact care plan that maybe an AI system suggested of like things of radiology or, or of, why am I thinking missing a drug

FRANKIE:

The chemotherapy.

Sal:

Yeah, chemotherapy is what I was looking for. Like three, rounds of chemotherapy, but the AI says, oh, you should do four or five. And that person, something happens to, it's like, who pays for what and who's at

Colleen:

Yeah.

Sal:

that.

Colleen:

And I know for a fact that, they're already making predictive decisions in the insurance industry on the other side, like home insurance and things like that. I think Frankie, you've had some experience with that. We've also heard stories recently of. Insurance companies dropping certain coverage in certain areas of the country. So I'm thinking about all the wildfires that were in California. Do you we think we're gonna get to a point with AI being part of that insurance industry where people really start to see kind of loss of coverage because they're better able to predict some of these, I don't wanna say catastrophes necessarily, but storms or wildfires or floods and things like that, I, I just kind of wonder how, I think it's just gonna affect everything and it's right now a bit unpredictable to know exactly how that's gonna affect everything.

FRANKIE:

Yeah.

Colleen:

Why don't we circle back to our overall topic today, about AI taking over jobs. I know we. Each really liked this article by Jack Kelly in Forbes. He was talking about what jobs will fall first as AI starts to take over some of these tasks in the workplace. Sal, did you have thoughts about that article?

Sal:

Yeah. So one of the main things is it's really kind of replacing some of the entry level jobs, some of the internships that people are taking. Those really junior level, legal roles, entry level IT roles, and it's really having a major impact in that. And I think one of the things that concerns me as more maybe a more senior person in this industry, is that's the next crop of people coming up and getting trained. And so who's gonna help us? With some of the additional work that like when it becomes more advanced stuff that AI maybe can't do or do as well, who's gonna have the insights from the next entry level people that are coming up and getting their skills

Colleen:

I think you make a good point. You know, these entry level jobs right now, we're used to having like younger, new people come in that we can then sort of train on maybe some of the things that are more task intensive. So that allows us more senior people to kind of move up the chain and to focus on, on things that are maybe more. Need our time a bit more. And maybe the answer to that, and kind of sadly, is that we use AI to do those more, not invasive tasks, but those more kind of putsy tasks. Those ones where you need to sit and write code or the one, those ones where you need to work on automating a process in some way. How do you think the next generation of like workers in the data space, let's just use that since we all work in the data space. How do we think that that's gonna affect people? How do people get into this industry going forward? If we are using ai, maybe in place of some of the work we would normally kind of use to introduce new people to, working in this industry.

FRANKIE:

I think it's gonna be as like somewhat hard for college graduates, but also at the same time with, you know, they're coming out of education, they're eager to learn. Most, most of them, they're, they're ready to go and they're like excited to get started. And I think that that will play a huge part in making them like. in hopping into career paths like this. So it might not be that they have, they learn these mundane skills anymore in college. Like that might adapt as well. They might be learning more advanced skills because they don't need to learn the the mundane, like I think about, I had a couple of classes when I was in my MBA where we were using AI and we were allowed to use AI to do our schoolwork. so it took away those, the mundane tasks, right? Like. When I was in marketing, part of my project, I used AI to just get some creative ideas of like, what could I do? I didn't spend as much time, maybe like critically thinking on some of that stuff. I, like, I just used the tools that I had in my hands and was able to use that to make my decisions. So it just might. Mean that colleges are gonna have to evolve so that those skills can come naturally to the students and make them more ready for, for those jobs.

Colleen:

Yeah. And that's interesting'cause I've had many, many conversations, in the last, six months to a year with people who are struggling to do that even now. It takes nine months to a year to come up with the curriculum, run it by the school board, and all of the powers that be either your, district or your university and to get that approved. And by the time that that curriculum is approved, things have changed. And I think that's why we see, even right now, some of the people who are coming outta college with these new degrees kind of struggle because they're learning things like Python, but beyond that, they aren't learning. Sort of those more advanced. Skills, and I think it's, we're kind of going through this growing pain phase, and I think we'll probably continue to do that for the next several years, until maybe they, they do start to teach some of these AI tools in the university, for example. I think about it too, like, you know, these kids are, they, they're learning Python because it's pretty universal. It doesn't really make sense to train them on any analytics or data engineering tools because those are gonna be specific to whomever your employer is. I think too, like a lot of that Python will be automated by ai and they won't need to write as much of it, but rather need to think of things. Kind of to your point, Frankie, like what you did, you know, think of like, how can I use AI to do this thing? How do I use these tools that have AI elements in them to get done What I maybe would've learned to code just, in Python in the past? I think that's conversations we just aren't gonna continue to struggle with until we sort of figure it out.

Sal:

From like a data perspective, I think there's gonna be a shift to people learning like what I'm calling data theory of how data moves, how to manipulate data, how to use data. And less of, Hey, how do I use, how do I learn Python? How do I learn, Tableau? How do I learn Alteryx? Less of the tools that you use, but more of like overall, how, how does data need to be moved and manipulated to maybe even use better in these AI systems framework for training of these models, better. So that information is getting. Paths to these models better. I think that's gonna be really important. And I think that's where the students and schools are not quite there yet, but I think that's where the students

Colleen:

Yeah.

Sal:

to go.

FRANKIE:

Yeah. That's kind of funny that you mentioned that. That because again, like I, I did that emphasis on analytics and so part of my schooling was like critical thinking about data and thinking about AI and things like that. so I think it's actually like moving in that direction and it's moving more than you, or maybe quicker than you realize. I spent a lot of time thinking about, or writing about like. How we were going to utilize data in our careers and things like that. And a lot of it was just critical thinking, and I think that's the skill that they're really honing in on, at least in, some of the smaller programs where they've been able to push through changes in curriculum quickly. I think that been a change and something that they're pushing for. The other thing that they were doing is, you know, like they're bringing people in. Who are working in these careers today, I thought that was really beneficial as well. Like hearing about somebody that is speaking about AI and how they're using it in their jobs and things like that. That was also super helpful just for getting me to start thinking about, okay, can I do that too?

Colleen:

Yeah, there's an article from, Harvard Graduate School, that talks about needing to know, if, AI won't take your job, if you know about IA and IA is what they're calling intelligence augmentation. So I think that's kind of what you're describing, Frankie, need to be able to shift from learning how to code the thing and sort of know how to theoretically use the thing and know how to apply that in a work setting. What do you think of that, Sal?

Sal:

That article also talks about. Critical thinking, collaboration, creativity, and are like some of the most important skill sets that you can develop

Colleen:

Yeah, I.

Sal:

I think that's completely spot on, especially around where that, that framework is going or whatever. For Harvard, usually if Harvard does it right, a lot of other schools will

Colleen:

Yeah.

Sal:

But I really do think it's really interesting. And part of this article also talks about. Embedding AI tools in, KK 12 schools so that they are learning these skill sets and how to use this technology earlier than we could ever have

Colleen:

Yeah.

Sal:

And so gonna be coming out with skills that maybe we can't even fathom right now

FRANKIE:

Right.

Colleen:

Right, because a lot of these things were new concepts, right? Like we would've never dreamt of this stuff when we were in grade school. But, judging by the way, at least for my kids, when they were going through school and how they introduced algebra and algebraic concepts much earlier, it was no longer a foreign concept to them by the time they got to the point in their, school days when they were actually learning. Algebra I, I think the same thing's gonna happen with ai, that they're gonna start to sort of drop these little hints, these little seeds, and I think it'll be to their benefit, these kids benefit to sort of learn about that and have it not be such a foreign, strange concept by the time they get to a point where they can do things with it.

FRANKIE:

Yeah, I remember like when I was in school, even just learning like how to Google and how to navigate the web was a big thing. Probably not so much anymore, like, because kids just know how to do all that stuff now. That was something that I can remember, like they assignments for us when we were in like end of grade school for like how to navigate and like, we had to figure out how to answer questions through like googling.

Colleen:

Yeah. You know, I recall my kids having to put together presentations and like, it's not Google Docs, but whatever the PowerPoint version of Google, is, around the same time that they were learning those skills, there was a big kind of flow bug in this country, that kids were no longer being taught cursive. And when you get over your shock over something that you thought was such a standard thing, you're like, yeah, you know what? I don't really use cursive. Who cares if the kids don't know how to, write in cursive, they're actually learning skills that they're gonna use on the job, or at least further along in their educational career. They're probably gonna need to know how to put together a presentation. Electronically more than they'll need to know how to write in cursive. I think we're gonna probably find other things like that where we are like all appalled for five minutes and oh my God, they're not being taught how to blah, blah, blah. And it's because really they're learning something at a higher level that's gonna be more applicable to whatever future skills are gonna need to build. Al you've got younger kids. Do you have any thoughts on that?

Sal:

Yeah, like there, this push or balance I guess between like giving him technology, but maybe not too early because yes, we have an instant gratification with technology, social media and stuff like that. And I think you're gonna also have that from a. Knowledge base from AI is like, Hey, I can literally ask any question and it will tell me everything I need to know about a, a topic. Right? And so I think there's going to be that and it's like how do we balance, Hey, there is, it is good to just take in information, but you have to have that, that other side of it of like. Hey, how are you critically thinking? How are you interpreting this information and making sure that it's right and validating it, and actually doing your own research to help with

Colleen:

Yeah.

Sal:

I think that's gonna be really important. My kids are so young that I don't know how

Colleen:

No,

Sal:

but I

Colleen:

but I think,

Sal:

my

Colleen:

yeah, I think you're my link though, to seeing how this all changes in the next, you know, 10, 15 years and how that becomes different in education. And we will have those clutch, our pearl moments where we're like, oh my God, they're not teaching, blah, blah, blah.

FRANKIE:

Right.

Colleen:

think it's gonna be really, hopefully most school districts, you know, kind of, latch onto these topics in successful ways that are helpful and useful, to start laying this groundwork earlier because as we've, been reading in preparation for this conversation, it's gonna be very necessary to be able to embed a lot of this knowledge into not just your work life, but probably all areas of your life.

Sal:

I was just thinking like way back when, like remember

Colleen:

Mm-hmm.

Sal:

it came out, and it was like the sign in it would take. Five minutes to sign in. You just hear the,

Colleen:

Oh, yeah.

Sal:

the dial tone constantly and now like how long it took us to connect to the internet and connect

Colleen:

Mm-hmm.

Sal:

I'll say to what our, like my kids will have is like they'll be taking in so much more information and it's really under to teach them how to

Colleen:

Yeah,

Sal:

that there's just so much

Colleen:

And this is completely a side note, but do you remember the early days of the internet, Sal? I'm sorry, Frankie, you're probably. A bit young to remember that. Maybe you do. I just remember when the internet was new and you had like these websites like hamster dance and literally the whole purpose of this website was to watch these little cartoon hamsters dance to goofy music, and you had all these really fun things because you could go to the internet and find cats. Um, that looked like Vikings to some Led Zeppelin song. Um, I remember all that very well and everybody had fun with it because it was the internet and you could do these crazy things and now it's like a cesspool of like dark, sad things that you could find on the internet. I feel like we're in that infancy with ai where people are making singing babies. Like I saw it was like, oh God, I forget the concert when I was a kid. It was like, we are the world. We're all the different, singers came together to, to record this music for some sort, USAID or World Aid, something like that. Yeah. And I saw an AI version where somebody, remade the video, but as babies. So there's like a baby Michael Jackson singing and a baby Kenny Rogers. And it was just, we're in that fun part of AI right now where people are making babies, singing songs. And I think too, we're probably gonna hit that point with AI where it's like, oh, now it's doing. Really bad, scary things, but

FRANKIE:

I

Colleen:

for right now,

FRANKIE:

there already because I think about like all the misinformation or I'll see something and

Colleen:

true.

FRANKIE:

oh my gosh, like what is that? And. Like Clay will be like, that's not real.

Colleen:

Yeah.

FRANKIE:

know, it's, sometimes it's hard for me to even decipher what's misinformation or like a, a generated image. and I think that's something that people are gonna have to learn too, is, and kids too, is what's real,

Colleen:

Yeah.

FRANKIE:

not.

Colleen:

Yeah, usually you look at their hands and if there's a person or supposedly a human, usually they have like six fingers or something or four fingers, and that's how you know that it's AI generated. But that may no longer even be the case, as I'm saying these words, they may have, it may have gotten better at that now,

FRANKIE:

Yeah,

Colleen:

but you're absolute.

Sal:

So.

FRANKIE:

too.

Colleen:

Yeah.

FRANKIE:

are really hard to decipher.

Sal:

so kind of to get back to our topic of AI and jobs, like as you, like, you have Colleen, you have young adults, that will be working in the next five years, right? Getting out in the industry. Like, are, have they raised any concern on like, Hey, I think AI. It's gonna be harder for me to get a job or how maybe how their career path or their decisions in school or

Colleen:

Yeah. That's a really good question. They are, young adults in college or about to be in college. I've had very few conversations with them about it. My stepson will be working in the medical field, not a doctor, but in labs and stuff behind the scenes. He's extremely smart. He's talked about, having. Topics introduced in some of his classes already to introduce these ideas of having AI give you, a first pass at your results, things that you would normally read. They put together slides, where they're analyzing cells and reading labs essentially, using AI or introducing the potential of using AI as a first pass there. And I think as far as we've. Discussed it. He sees that as being part of, something to aid him, like a tool in his job. I'm sure that'll probably change even over the next two years as he finishes his degree. But, I think most of them, see that AI still as being something that can be kind of cool, that can assist you in what you're doing with your career, but not necessarily anything to be really afraid of.

Sal:

Do you think that AI is plagiarism? And the reason I ask this question is'cause like schools and a lot like, oh, don't use AI to write your papers. Don't do that. But like at some point, might switch to be like, it can write your papers, but it can't directly pull like

Colleen:

Yeah.

Sal:

or from articles and stuff like that. Like I wonder if there, there is this balance there, but I, I am curious if your, what your thoughts are, or both of your thoughts are, and is, is writing using AI to write articles and write things for yourself? Plagiarism is it. Should it be negatively looked at or should it be positively looked at that,'cause the reason I say this is companies and in research at schools, it's going to change how people use it because if, if it's negatively impacted or negatively looked at, I think people are gonna be more hesitant to use it.

Colleen:

Yeah, I have two thoughts on that. I do think that our perceptions of how we use AI are going to shift. I personally, like I will use something like chat GPT if I'm making a. Post on LinkedIn, and I really wanna get people's attention. I'll write it up, and I'm somebody, again, I have a writing degree. I think I communicate very well. I'm still usually more impressed by what chat GPT comes up with than what I've written and think it's more concise and kind of has the voice or the feeling behind it that I was going for. So I think that to a certain extent, the teachers who say, don't use. Chat, GPT or don't use AI to write this. There may be some shift there, but the question is to whether it's plagiarism, I mean, think about what is being used to train these models. If you're taking people's written work that they have created themselves to train a chat bot, for example. Maybe there's a literary chat bot that somebody creates to help people write papers. I do think that the people who wrote those original works. Need to be compensated or even credited for helping to build that. And I think right now a lot of those works are not being, attributed in that way. And I think that that sort of needs to be remedied. And I think you should be able to opt in as to whether or not you want your work, whether it's art, whether it's, something you've painted or sculpted, some photo that you've taken, whether it's a book you've written. I think those things need to be, cited like as you are the source of that and you need to be able to determine whether or not you want your work product to be part of training, whatever future models that that is used for. Frankie, do you have any thoughts on that?

FRANKIE:

Yeah, I was kind of thinking of right along the same lines. It's like anytime that I had to write a paper, what I had to do was cite the source, right? Like I was pulling information directly from all these different articles, quoting it and then sourcing it, I think that as long as you are doing that, that it's okay. And it's not plagiarism. But I do think that the chat bots and like chat GBT needs to get better about sources.

Colleen:

Yeah.

FRANKIE:

done some improvements there and they've been listing articles and things like that. Oftentimes after I ask a question. But they do need to be better about writing out, like where did they get that information? Because I. Might wanna dig through it, I might wanna go back and see did they pull this out of context or did they pull it correctly? Like,'cause sometimes it doesn't have that human nature, knowledge to it. So being able to dig up the source and figure out exactly where it's stemming from is super important and needs to be, an enhancement on all the chat bots.

Colleen:

Yeah. I wonder if that's like a prompt engineering thing. I wonder if you could like ask chat GPT and I just, I'm using that as a generic term here. Just, just know I'm using it like Kleenex, right? Like if you could go to a chat bot and basically ask it a question and please cite your sources. I wonder if it would do that. Like, here's an article from Forbes, or here's a article from somebody's blog that I used as the source of this.

FRANKIE:

Like after I've put a question out there, where did you get that information? And it'll tell you some sources, but I think that maybe it's not hitting everything, and that's the part that I would love to get more information on.

Colleen:

Yeah.

Sal:

I am dyslexic. I think I talked about that in my, behind the mic. Helped me tenfold'cause I am not a good writer. But I feel like I have a lot of information that I can express and things that I've learned throughout the years that I've worked, uh, that documenting it and making sure has be AI is, I've used a ton to build out that documentation, truly been amazing for me. For people like me where I do have a challenge of writing, I think it's gonna be a massive help to individuals that have the skill sets, have the intelligence to build and create and think new ways. And I think this is where the jobs will start, I have a way I want to create an app and I have this idea, this entrepreneur spirit, but I don't know how to code and I don't know how to do all these things. So I'm gonna go out and just use chat bots to really build what you consider as like a MVP, like a minimal viable product and then get better coders or more expensive AI systems to help build that out. That's where I think this is gonna go, people are gonna become their own much faster as they have ideas and build out their own businesses, using this versus working your way up at an industry. So I do think those entry level jobs are gonna maybe diminish a little bit, but there's going to be other opportunities out there

Colleen:

Yeah, let's hope. Sal I wanna bring up something. You and I kind of had a side conversation about some of this the other day that I thought was kind of interesting. We came up with some interesting points. We were talking about some of these articles where you've got CEOs of really big tech companies coming out to say some of these things, to basically saying, Hey, AI might wipe, wipe out half of all entry level white collar jobs like. You know, what do you make of like the CEO? Was it the CEO of Amazon, coming out to say, he just says the boss of Amazon, but, coming out to his workers and therefore publicly to the world to say, Hey, AI is gonna wipe out a ton of these jobs. What do you make of that, of them making this public declaration like that?

Sal:

Yeah, I think when we were talking about it, I was like, is this a real fact? Because the investment that we're gonna do, we're gonna change our hiring strategy to reduce, the white collar workers or the entry level workers, and to match that 60%. Or is it a

Colleen:

Yeah.

Sal:

move? Right. Is like get on board. Start using these systems. If you're not using these systems, you're gonna be out. And then second is, hey, now you have a really productive worker that you're being now benchmarked against, now you have to

Colleen:

Yeah.

Sal:

And so I'm like, where is he? Where is he? What was the tone that he wanted to do this? Is it, is it to put pressure on his employees to be better employees or to. into, Hey, we, you need to upscale in these things or else

Colleen:

Yeah. I mean, and it is just not, it's not just Amazon either. Dario Am Modi, the CEO of Anthropic has said similar things as well, and I almost wonder if they're trying to create like a self-fulfilling prophecy. To your point, Sal, like, are they saying these things because they've invested so much in the AI industry that they wanted to become? Facts, you know, sooner rather than later. And to your point, like we know AI tools can really increase productivity. You no longer need another worker to do all of that programming for you or to create something for you. You can have an agent do that while you then do other things. Things get done, you know, the, the, A agent can complete that work much more quickly. And now you've got a workforce that is being pressured to step up their game and they're producing 1.5 to two times what they normally would like. We were sort of laughing about it the other day as we were talking about this, but really that could be a way to drive this innovation and sort of speed it up and get it to, take place sooner rather than later.

FRANKIE:

that's a really good point. I had never really even thought about that. They think they might just be like saying these things to try to use their influence and make things happen.

Colleen:

Mm-hmm.

FRANKIE:

I thought might be like trying to reach their shareholders to let them know like, oh, we're gonna be

Colleen:

Yeah.

FRANKIE:

so we're gonna make more money, and you know, things like that.

Colleen:

Right. We're gonna be more productive, have lower overhead.

FRANKIE:

Right. But then, you know, like in that Forbes article, they had referenced that McKinsey reported that by 20, 30, 30% of current US jobs could be automated and 60% significantly altered by AI tools. Thinking about that from a third party perspective, right? Like they don't have any stake in a company like Amazon or Anthropic, right? And in theory they don't, they shouldn't, but.

Colleen:

doesn't?

FRANKIE:

the McKenzie

Colleen:

Oh yeah. Institute. Yeah.

FRANKIE:

So I just think if they don't have any stake in that, why would they be spitting facts like that as well? And McKenzie, I feel like is always a little high on their ranges. I don't think that 30% of jobs will be automated by 2030. seems a little excessive in my opinion. That's only. Four and a half years.

Sal:

This is where they're probably looking at

FRANKIE:

Yes.

Sal:

jobs, right? They're not saying, Hey, there is gonna be. 10 x of jobs that were created because of this, right? They're like, oh, of the current jobs that exist right now, 30% of it's gonna be automated, but we actually gained, when I say jobs like. of industry or like, positions, not actual like individual jobs. I think we're gonna have a massive gain in like AI engineering and all these people that are actually setting up all this, this automation and how to develop it at multiple size companies. I think, like, I don't think a company can invest millions and millions of dollars

Colleen:

No.

Sal:

startup, right? And so like, there's gonna be industries that are. That are going to need the human to

Colleen:

Yeah.

Sal:

of build this, and figure this out. And I think those jobs are gonna be growing. So I think there's this balance between, yes, we might have a current, looking at our current jobs, a 30%, automation increase, but

FRANKIE:

but like I think about like societal change, it could happen really fast. Yes. But like when we had to start ordering on QR codes at restaurants as humans, did not like that. Like a lot of the people, you go to a restaurant and like expect to be taken care of, right. And

Colleen:

Yeah.

FRANKIE:

some restaurants are still doing it and they're, they're having success. But a lot of restaurants that did that went back to having servers.

Colleen:

Yeah.

FRANKIE:

that AI change in like some of the jobs that they're referencing are communication positions because they think that can take over scheduling and things like that. And it just requires so much more of a society change than I think what they're kind of calculating for. I don't think people are gonna be super on board with that. Especially like thinking about like our older generation. They're specifically the ones that I've seen the most, resistance to change with, it's not even like, they're not even that old of people. Like I think about like my parents, they don't wanna have to call to schedule and hear a chatbot. They absolutely hate it. Like

Colleen:

Yeah,

FRANKIE:

you know,

Colleen:

yeah. No, no. Honestly, I, I was actually thinking I'm probably not that much younger than your parents, but my absolute pet peeve don't, don't make me download another app. I don't wanna do that. McDonald's, I'm not, I'm not downloading your app. Sorry. I'm not gonna do it. My insurance company not downloading your app. I can probably go through my website and get all the information I need. I do not need more apps. That is my old person pet peeve that I won't let go of. Yeah, I feel like too, there's another side to this. These CEOs coming out to say these things, they probably own stock in their own companies, so it's kind of like. Lining their own pockets to come out and publicly say these things and to really nudge their employees into making, change sooner rather than later. Right.

FRANKIE:

absolutely. What are you guys doing at work to enhance your productivity with ai?

Sal:

I think what I said earlier is like I'm able to, there's a couple things. So one is from a documentation perspective, I'm much more efficient. In writing up documentation, filling out, user manuals and stuff like that for people to use. Which I can then focus on more of my time on what I consider data theory is more of, Hey, how does data move from end to end and build out, a process to get to an end result or a, data-driven decision. So you're getting to that point, quicker. then within like coding perspective is because I know that data theory, and again, I'm using this my go-to kind of term for it. I can now structure my like coding prompts to help me code and get to an end state much faster and a. at least a framework or a structure of that I can then take my knowledge of, Python and add additional coding to it. And so I can move way faster than I've ever had

Colleen:

Yeah, I definitely have seen it automate things that I didn't wanna have to do before. You know, to your point, documentation, I've had, some AI agents do really well at creating that documentation. One I've been specifically loving lately. There's an integration where I've been working that you can take a message sent in a Slack tool for example, and you can use that to create a ticket so that you then have a issue on your board to then work. And I love that it fills in most of the boxes for you. It's usually pretty accurate. It usually includes more information that I would probably include if I were manually creating that ticket. And so it's those little pieces of automation that just sort of make the workday easier and faster and get you through some of those, like really putsy. Tasks that you don't wanna have to do.

FRANKIE:

Yeah, and for me, I would say like sometimes I have to do some company research and instead of having to read a whole 10 K report on a public company, I just ask for, you know, a summarized version from Gemini or just ask it specific questions relating to that particular report so that I can, instead of having to read through it and find it, I can just read like a quick couple of sentences. So that's been really helpful for me.

Colleen:

Yeah.

Sal:

You guys will laugh at this, so I actually created a AI podcast research assistant. Agent. Yeah. And so I give it the articles. It helps me summarize all the articles, think of new questions to possibly come up with. And so it helps me do this podcast. And it's really easy, and structure it. And I can reference information and I don't have to put in the significant amount of time to kind of

Colleen:

Yeah, I know I've used it in the past in prepping for some of our interviews to say like, if we were interviewing a person in this field, what should I ask?

Sal:

Yeah. Kind of like, again with the assistant and I think where the world's gonna go, and I think where you're gonna see significant job replacement or maybe job enhancement, but I could see job replacement is this, agent, perspective is start as we start to move into this agent and what it's called, like MCP, which is, model context protocol. But like is where the LLM or the, the chat if you're not familiar with the LLM, we'll start doing tasks for you. And that's where I think it's like, oh, okay. As soon as it starts building and being. An analyst or doing something for you, or for an individual. That's where I think we're gonna see job replacements around like, especially around like maybe secretaries or something of that sort. You could build an AI secretary that you could say, all right, document or update my, outlook whenever a new email comes in and has an invite, and tell me a summary of what we're gonna be talking about. Like what secretaries used to do, I think in AI

Colleen:

Yeah, and I've heard personal stories and read about many instances where people use, AI agents to replace or enhance their customer service desk. Whereas before they may have employed five people, they may now only need one because 80% of the questions can be answered or have solutions provided by the AI agent, and you maybe just need a person on hand in case something, falls outside of what can be handled 80% of the time.

Sal:

So even for that example, right? You wanna return an item or you complain to a chat bot. Right I think what it's gonna have end up doing is it's gonna take in that information, digest it. Create actions from that and maybe send you a new product, maybe send you a coupon, a discount card, like where that would've actually take in a human interaction before of like, oh, now we gotta go fill out, an order from our supply or our distribution group. And then that goes out and gets it shipped out, right? Like this is all gonna be done by an

Colleen:

and it's really more agentic, right? Like what you're describing. Whereas before you would go and maybe your company's got like a little portal where you can ask a question, where can I get my HR forms for blah, blah, blah. You know, that's kind of where the chain dies. Agentic AI then takes that a step further and can troubleshoot that for you and says, would you like to change your tax withholdings? And yes, I would, and here's what I wanna adjust it to. And then that agent goes out and makes those changes for you. So I think that's where the shift is gonna come next. More from this kind of, yelling into the void, asking these questions of this kind of. Quote unquote robot behind the scenes that just spits out an answer into something that can then chunk that into pieces. And to your point, like maybe we need to ship out a replacement part, right? You go down this decision tree, you answer all these questions, and the AI agent spits back out your new widget will be arriving on June 30th.

FRANKIE:

The best part is like, it doesn't need to be prompted to do the job.

Colleen:

Right.

FRANKIE:

that's, the huge difference in, agent AI is it can just do it without you telling it to go do it.

Colleen:

It doesn't need a human person sitting behind the scenes to say, oh, they had a part go bad. Let's ship out a new part. Now I need to contact the, the mail room and have this. Parts shipped out or whatever, that a agentic agent can then go to, inventory, make sure that they have a widget in stock, fill out an order form for this widget, say that it's a replacement part, have it, go to the shipping system to have a label created for this box that's gotta go to ABC box company or whatever it is. I think there's some really exciting things happening there to think that we could automate all of that. I think we could probably have a whole episode just talking about that.

Sal:

One of the things with like IIOT devices, so internet of things, a lot of things are connected to it. So let's say that you have an air conditioner in your house, right? And it's connected to your system,

Colleen:

Mm-hmm.

Sal:

or to your wifi, and it connected the internet and it talks back to the builder or whoever, like Elegy or whoever built your, your, air conditioner and then it tells you not just maintenance, but how is the health of it. And then it has an agent on that side sending you out new filters and new things at different schedules based on what the health of your air conditioner is. Right? I think that's where it's gonna be like, holy cow.

Colleen:

Yeah, like could you imagine if you had like a big manufacturing company and you've got this production line? Well, if there's pieces in that production line, again, I'm gonna say widgets and sprockets, I can't remember what they're actually called, but you've got, physical machine parts that are moving your product along. If you've got IIOT sensors there and you know that this particular widget is going bad, it's got maybe 20% of its life left. It's supposed to last 10 years. It's only been, eight. If you can have a replacement part shipped to you before that part actually goes bad and stops working, you can avoid any potential downtime there and you can have your maintenance people replace that belt or that piece of that. Conveyor belt before it actually breaks and avoid any kind of downtime. I think that kind of thing is really, a really cool application for things like this. So do we have any final, parting thoughts on AI replacing jobs? I mean, I feel like a lot of people might listen to this and think, oh God, like I'm never gonna get a job in this industry now, or I'm never gonna get a job, period, because AI's just gonna come in and take over. I don't know if we have any great advice for new people coming in other than to say, familiarize yourself with AI and all of these concepts, and kind of just prepare yourself for things maybe not working the way that they always have. Does anybody have any other advice for new people coming in?

Sal:

Yeah, I would second that is like, start to learn it now, understanding. I mean, it's moving so fast that it's hard to keep up. Don't think that you need to keep up, but just know that, just start understanding the concepts of it all, and how it's gonna be used and how you can use it. I think we're going to have this shift. I think we are gonna have some job loss, but I think we're gonna

Colleen:

Yeah.

Sal:

some job gains. I think there is some positive

FRANKIE:

and it doesn't need to be quick either, like just. off with, you know, a chat GBT or something. Ask it a question and see how it responds and start to think about, you know, what other questions could I ask and could this actually help me in my job today?

Colleen:

And I don't think anybody needs to you like, you don't need to learn the programming. You don't need to know how to create a bot or an agent of any kind. Just get familiar with using it and how those concepts might be applied in the workforce.

Sal:

Well, that wraps up our show. We had great discussion, on ai, and we'll take our jobs, as always, subscribe, follow us, share it, on any of your social medias.

Colleen:

And until next time, keep calculating.