Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
The token math ain’t mathin’, so time to get back to what makes us human
123 Plays9 days ago
The hype machine spent two years telling us AI was coming for your job. Now it's quietly walking that back. Why now? Follow the money.
On this week's system update, George K. and George A. pull apart the vibe shift happening at the top of the AI economy: from Uber's COO admitting he can't draw a line between token spend and shipped features, to the broader reckoning hitting every CFO who signed a three-year AI contract without modeling what agentic workflows actually cost.
The subsidized era is over. The bill is due. And nobody has a clean answer.
But the harder question underneath all of it isn't economic. It's human.
What happens when an industry skips straight from "how big can we make it" to "what are humans even for" without stopping to answer either?
The two Georges reckon with soft skills being repackaged as vital skills, the neoliberal bargain sold to a generation of college graduates, and what Pope Leo's 42,000+word encyclical on human dignity in the age of AI gets right that most boards and governments haven't.
A tech podcast about humans. This week, more than ever.
Mentioned:
- Jensen Huang on irresponsible proclamations
- Uber COO on lack of ROI from tokenmaxxing
- Ed Zitron on OpenAI and potential collapse of Oracle
- Daniela Amodei on the importance of the humanities
- Jamie Dimon on future job skills
- What 2026 hiring managers are looking for
- Pope Leo XIV’s encyclical, Magnifica Humanitas
- Marissa Alert on business outcomes planning first
- David Homan on how to build real human networks
- Sharon Goldman on the small town impact of the datacenter buildout
Recommended
![The Everything Machine and the Trillion-Dollar Bet [Replay] image](https://media.zencastr.com/cdn-cgi/image/width=112,quality=85/image-files/6322f03255d97a700020586e/f0ee1901-e926-4895-9598-4c323bd806cb.jpg)
Transcript
The Rise of Soft Skills in Tech
00:00:00
Speaker
The MBAs who spent a decade telling kids learn to code philosophy is luxury are now on LinkedIn, quoting Aristotle and posting about emotional intelligence. I know. So fuck those guys, by the way.
00:00:12
Speaker
read Read the room before you read Plato, folks. And I actually, I called this out on a post actually, like I think a month or two ago. um And I have been talking to people about this. It's not a coincidence since the pivot lands the same year their headcount math broke.
00:00:25
Speaker
Right? Soft skills are becoming the new gospel precisely when the hard skill or placement story stops working. you know And and a cynical read of of soft skills is a convenient frame for what we um we don't actually know. or on. Let me think of how to word this. We don't actually know what we need anymore. so let's just call call it judgment and move on. Right? 100%. 100%. The humanities kids weren't wrong. They were just early and broke.
Introduction to 'Bare Knuckles and Brass Tacks'
00:01:02
Speaker
Yo, this is Bare Knuckles and Brass Tacks, the tech podcast about humans. I'm George K. And I'm George This week, y'all, we are going to do a system
Economic Realities vs. AI Promises
00:01:12
Speaker
update. If you are new to the show, we typically have a guest interview, but occasionally we take time out to just catch up on all the shit that's going on and reflect on some of the conversations we've had. And so we're going to do that today. All right. So we're going to start out with the vibe shift.
00:01:29
Speaker
um Turns out all those companies that were loudly proclaiming that AI would eliminate all the jobs and automate all the work seem to be quietly walking that back and some divisions within the magnificent camp.
00:01:47
Speaker
All together, you know, i had Jensen coming out and saying that such proclamations are dangerous and irresponsible and the fact that NVIDIA is actually hiring more engineers. um So just get your thoughts. Why now? what's What's going on? I got a few hypotheses, but what do you think about that?
00:02:04
Speaker
i mean like i I look at it. 18 months ago, every CEO was on stage, you know doing the layoff math out loud proudly. Now the same mouths are pivoting to human AI collaboration. Like, like nobody saved the tape.
00:02:20
Speaker
um I think the walk back isn't a strategic update. It's a narrative damage control. know The models didn't get worse. The economics just showed up to the meeting since now it's no longer subsidized to get all these tokens used.
Financial Pressures on AI Spending
00:02:34
Speaker
ah Klarna already did the round trip. I mean, they fired the humans, replaced them with AI, then quietly went back to hiring humans um when the customer experience like basically cratered and everyone else just learned to skip the step one in public, but they all went through the same, like ask anyone that had has anything to do with working with like Salesforce or Meta or countless other companies.
00:02:59
Speaker
Uh, that's, that's kind of it. Um, don't know, like, I think a question now we have to ask is like, why now? Like, is this a genuine reassessment or is it, uh, you know, IPO pressure management? Like we kind of been talking about this and, know, we kind of, you follow the calendar. I think we're inside the window where, you know, the 2024 vintage AI promises, um, have to start showing up in earnings calls and the gap between the deck and the PNL is getting hard to, to caveat. um You know Uber came out with that statement and recently with their COO saying that they blew through their budget for the entire year for AI token. Yeah, so that that was a really public moment of hilarity, right? So their CTO does an interview with the information and says they blew through their Claude Code budget in four months, which caused a meltdown inside Uber. And then the COO comes out and is like,
00:03:53
Speaker
Look, man, I cannot draw a clear line between our token spend and the number of features we're shipping, which for a software development company is pretty stark.
00:04:07
Speaker
i mean I mean, look, man, if we're looking at, if we're looking at what the honest answer is, i think I think the problem comes down to some of these executives are genuinely seeing the limits up close and the smart but smart ones know they need to find an exit ramp before the analyst questions start getting pointed. Right?
00:04:29
Speaker
Yes. I do think though, when the messaging starts to shift in unison across competitors, that's not just a reassessment. That's a peer group reading the same Goldman note and adjusting their talking points before the next investor day.
00:04:43
Speaker
And I think like,
00:04:47
Speaker
Look, I've always said workforce guy. So are you. It's the same deal. Sorry, go on. The market is vibes, right? Like economics is, can be studied in hard math, but a lot of what happens out in the world is based on feelings. Like it's a legitimate metric that has been ignored. And so if you have a whole bunch of people saying the same thing,
00:05:13
Speaker
But then the vibe shifts among the investor class who are like, wait a minute, you're spending how much on whatever? Wait, what will that do to shareholder returns? And they start asking these questions. And then that's public companies, to your point. If they can't
AI Pricing Models and Budget Impacts
00:05:28
Speaker
answer those questions, stuff gets very real for them, right? The likes of Uber and and banks and stuff like that.
00:05:35
Speaker
for the folks that are trying to IPO, I think there's also a culture clash, right? Like it's one thing to live in San Francisco and kind of live in the transhumanist, whatever bubble, the hyper-rationalist, effective altruist, whatever cultures,
00:05:52
Speaker
Heat Dome is there where you all kind of have the same language. Then you start going to big East Coast banks to bankroll your IPO and they just don't have time for that nonsense, right? Like there is a world outside San Francisco where they're like, how are you going to do this when you're, you've taken out this many loans to build this stuff? ah The math says you have to make like $200 billion dollars by 2030 or you're not going to be able to pay for these things, right? So that there's math on two sides. There's the pressure on the company itself. And then to your point, the people who have to live with it today.
00:06:30
Speaker
And so I think for our listeners, if you haven't been paying attention, The underlying economics here that George has mentioned is a pivot from inference, aka the results of the chatbot, being largely subsidized by huge funding rounds.
00:06:46
Speaker
And it costs a lot. It's very expensive. Ed Zitron has pointed out this is the only business model where your costs go significantly. commensurate with the amount of use that your highest users are using the tool, right? That's sort of a weird business model.
00:07:02
Speaker
So now they're shifting the pricing to token-based pricing, which is ah obviously token is kind of like an arbitrary unit of input into the model.
00:07:12
Speaker
And so if that starts to get very expensive for a company on, say, a Claude or an OpenAI codex, co-pilot, whatever like enterprise license, they are left holding the bag, right? So you said like everyone use AI all the time.
Governance Failures in AI Investments
00:07:28
Speaker
oh wait, I don't have anything to show for the bill becomes a hard business problem to crack.
00:07:35
Speaker
I actually did a little bit of digging into this and I want to provide like a bit of an executive perspective on this as someone that has to manage a high-end P&L. So, you know, McDonald, that COO, he said on the rapid response podcast, like in very plain English, um that link is not there yet.
00:07:53
Speaker
That was his quote. Now for a sitting COO to put that on tape is to tell that the internal numbers are way worse than the public framing. I'm talking like probably 95% of Uber's engineers are using ai tools like monthly. So more than, I'd say one in 10 lines of code is ai authored. And, you know, the chief operation officer still can't point to a feature and say that one shipped because of this. So the KPIs, the metrics aren't measurable. So going to a shareholder with that is not possible.
00:08:24
Speaker
The adoption is not the outcome. And he also flagged that the real problem, no one wants to say out loud, is that engineers don't pay the invoice. So they treat tokens like free water. Right. and That's not an AI problem. That's a basic governance failure that exists in like every shop right now.
00:08:39
Speaker
And maybe culture failure because like, remember, they were there were some companies that were basically saying like, if you're not token maxing, then we see you as
Economic Risks of Data Center Investments
00:08:50
Speaker
less productive.
00:08:51
Speaker
ah That's a whole other piece of insanity. But like, you know, you called it out. Like their CTO called it a quote, head exploding moment, which bro, your your head should have exploded like a year or two ago. um Nope. Like, here's the thing.
00:09:05
Speaker
Nobody at Uber modeled token consumption against the Gentic workflows. And they're not the only ones. Like most CFOs I've talked to about this have no idea what their actual 2026 run rate looks like. It's like it's like ah they literally treat us like it's not my problem, but it's directly their problem.
00:09:25
Speaker
Four months in full year gone a budget. That's 200% overspend trajectory if you straight line it. And it happened at a company with serious like FP&A a muscle.
00:09:37
Speaker
So imagine what it looks like at like a 500 person SaaS shop with no AI cost discipline. The deeper issue here, man, and this is really what I want to get your take on.
00:09:48
Speaker
agentic models eat tokens at multiples of standard inference, right? And the pricing model has already shifted to usage base. So Anthropic changed from flat fee to metered.
00:10:00
Speaker
Like Sam Altman, try and not to say they didn't get us sued. Sam Altman is openly comparing it to electricity billing. I think the era of predictable AI budgets is over.
00:10:14
Speaker
Yes. And
00:10:18
Speaker
I don't know how to say this without just sounding like an old man screaming at the clouds or stating the obvious. ah but But it's a probabilistic output,
AI Governance and Market Demand
00:10:29
Speaker
right? Which means the consumption is really high because you might spend a lot of time trying to tune the output because it's not like an if-then thing.
00:10:41
Speaker
deterministic output, right? So once the business output is a little squishier around the lines, you've got to do more work to massage it, whatever. That's why you're eating up so much in the token budget.
00:10:51
Speaker
And then now you've got to do a behavior change, right? Because like for the last two years, it was just like, you know, do whatever you want, free water, right? Oh, wait, now I got to pay for the water. Now it's a ah cost center to the business and I have to be able to point to a return, makes it a little trickier. And then, know,
00:11:07
Speaker
This extends out to the larger economics when we talked to Sharon Goldman earlier this month, right? If they are saying they're going to place this bet, $700 billion dollars worth of investment in data center are build out on something that the revenue is not playing out, that is bananas problem. We've talked about in that episode, but potential cascade of economic damage should data center building pop as a bubble.
00:11:33
Speaker
But George, i want to get your take because you run a ah shop that has SDLC and stuff like that. Everyone who has bought either security tooling, marketing tooling, sales tooling, any software purchases in the last three years that involve quote unquote AI, which is really either a wrapper or some sort of API call out to a foundation model, so-called,
00:11:58
Speaker
is also going to be affected by me token pricing. So does that cost not just then get passed on into your SaaS bills? Like the bills are going to go up because their bills are going to go up.
00:12:10
Speaker
Yeah, you know, what this is, this is, these are conversations I was literally having this week, like at work, um you know, with not only my team leads, with my architects, with our procurement leads, trying to figure this out, like, hey, like, this is like a ah problem I'm really foreseeing. it's something that eventually, they'll probably have to be a
AI's Societal Impact and Sustainability
00:12:28
Speaker
C-suite level conversation I'm going have to have be like, hey, we're going to have to turn the course on some things, which thankfully, like, you know, I'm i'm with my architects, I'm helping drive a whole AI governance infrastructure and framework conversation, which will, know,
00:12:40
Speaker
allow that sense of control since it ties everything to business outcomes. That's a separate conversation. But I think the issue here is that the hyperscalers are committing CapEx on the assumption that enterprise demand will absorb it. Right. So in the Uber case, their COO just told everyone that, you know, the demand signal is softer than the spend suggests. And I think that's a problem that you know travels upstream pretty fast. And we've seen this movie before, right? Like in the telecom world in 1999, fiber everywhere. And then 20 years of dark fiber, no one could afford to light. um you know And then the AI InfraBuild is running a similar play.
00:13:16
Speaker
But, you know, with a different ticker. And so the interesting question I have isn't whether the bill that it's too big, it's who absorbs the loss when the demand curve catches up to reality, right? My money says that, you know, it's the hyperscalers that are going to have to. It's the enterprises holding the three year committed spend contracts they signed in 2025. They're the ones that are going to pwned.
00:13:39
Speaker
Absolutely. And shit only flows downhill. um So i also think to, we did a replay this month with our interview with Dr. Marissa Alert. And I think this all comes back to what you had been saying for months now, which is if you just...
00:13:56
Speaker
use the AI for its own sake and no business outcome strategy. This is where you find yourself. You're like, do the thing just for the thing. oh wait, the thing now costs money. Damn. Right. Like there was, and there was no plan. There was no plan. There was just hype and breathless excitement and peer pressure, essentially.
00:14:13
Speaker
But these are the business leaders sort of just running like lemmings off the cliff. Right. Yeah. And look, we have thoughts to look at like, again, to our, to our tour interview with Ms. Goldman, I believe her name was. i Yeah. Yeah. Sharon Goldman.
00:14:30
Speaker
Sharon Goldman. I do, I get punched in the head way too much and I just don't sleep enough. So I forget things, but Sharon, you are awesome. Thank you for getting that interview. um You know, I follow a lot of, um a lot of other independent journalists. So James Lee is really good.
00:14:43
Speaker
And sometimes he gets kind of crazy because he gets a little bit sardonic in his post and stuff. But he also has talked about what are all these data centers? Like, what are they doing?
00:14:54
Speaker
Like, when you see, like, the amount of power and everything generated and produced, and you see the environmental impact, like, he recently put out a post which I... Thought was like half tongue in cheek, but it's like these things could literally be repurposed as concentration camps. And I'm like, that's kind of terrifying. But it also like you look at like what is with all this behemoth infrastructure that money has been wasted to to build.
00:15:17
Speaker
And now we're at a point on a promise that may not come true. And it's not. And i think I think organizations are just very quickly going to realize that they can't actually afford to play this game anymore.
00:15:31
Speaker
And now we have gone into a world where we are destroying our own environment to what ends. We are impacting the health and wellness of people to what ends. And, you know, we've created this condition now where we have in-shittified the entire economy.
Financial Sustainability of Major AI Players
00:15:47
Speaker
to what ends. I mean, I'm not trying to, to, to detract from what you're saying. I'm just saying that there's a much bigger issue at play where like yeah all this infrastructure has been built. What are we doing with it? What is, what now? Because no one's going to be able to afford to use what the purpose of it was to begin with.
00:16:04
Speaker
Yes. And and I will ah will close it out here on this part of the topic, but I would point people to Ed Zitron, who actually gets into all the filings and does the math and speaks to data center contractors on the ground. And um I don't remember the exact figures, but...
00:16:20
Speaker
If OpenAI cannot make a certain amount in revenue by 2030, they cannot afford to pay Oracle for the data centers they have been contracted to build. In other words, it is a big enough problem that the so-called too big to fail or the, oh, Oracle will never go out of business. Like there are things in this weird looping financial system that could take down some very heavy players. So... Not to be underestimated.
Shift from Technical to Soft Skills
00:16:50
Speaker
um Okay, so let's pivot here. We've talked about how we started with business leaders saying like, oh, wait, maybe not all the jobs will be replaced. But what are the skills that they do want now? So this started in December of last year. Jamie Dimon, of all people, saying...
00:17:11
Speaker
that while AI would automate some work that they would continue, JPMorgan Chase, to look for people with strong soft skills, the ability to listen, the ability to strategize across business lines.
00:17:23
Speaker
ah Daniela Amode, Dario's sister and president at Anthropics, as humanities will be more critical than ever. Hiring managers are listing communication and critical thinking and resilience as top 2026 hiring priorities.
00:17:38
Speaker
so Is this like a genuine reckoning? You and I have been trying to rebrand soft skills as vital skills for two years now, or is this just some new thing to say? Because, you know, it's graduation season.
00:17:53
Speaker
a lot of folks leaving college not with a lot of hope for the job market. what's your What's your take here? i mean, look, I think...
00:18:05
Speaker
First of all, the whiplash is something else, right? Yes. The MBAs who spent a decade telling kids learn to code philosophy is luxury are now on LinkedIn quoting Aristotle and posting about emotional intelligence. I know, so fuck those guys, by the way. Read the room before you read Plato, folks. And I actually, I called this out on a post actually, like I think a month or two ago. um And I've been talking to people about this.
00:18:29
Speaker
It's not a coincidence since the pivot lands the same year their headcount math broke. but Soft skills are are becoming the new gospel precisely when the hard skill or placement story stops working.
00:18:41
Speaker
you know And and a cynical read of of soft skills is a convenient frame for what we um we don't actually know. on, let me think of how to word this.
00:18:52
Speaker
We don't actually know what we need anymore. so let's just call call it judgment and move on. 100%. 100%. one hundred percent one hundred percent The humanities kids weren't wrong.
00:19:04
Speaker
They were just early and broke. And I mean, I'm a i'm a guy with a ah humanities degree. I studied politics and psychology. Same anthropology. like so Here we are.
00:19:14
Speaker
my my my My favorite things to read are are history and mythology and and and classical philosophy. Yeah. You know, it's, we, we didn't end up where we ended up in our careers because, you know, we were like technicians, like i'm i'm a capable technician. You are a very capable individual with technology.
00:19:34
Speaker
What got us here was our ability to see things both, you know, on the ground, looking up, at the forest.
Education's Role in Workforce Preparation
00:19:42
Speaker
And again, being able to step back and look at it from, you know, the 10,000 foot level and understanding, you know, what exactly is this landscape that we're dealing with and providing leadership to guide organizations in ah in a better direction, which then escalate our careers.
00:19:58
Speaker
I think again, you know, You look at like, what was it? You mentioned Danielle Amadei, right? Danielle Amadei, yeah. Amadei, sorry. you know She was saying like humanities will will will be more important than ever.
00:20:12
Speaker
I fully agree with that. I fully agree with that. And I think, you we talked about that previously too when when I started doing some semi-ag governance training months ago. And I was just like, so here the things that going to get automated out.
00:20:23
Speaker
And then here's the things where humans still need to remain. Our creativity, our sense of judgment, our ability to build decisions, our ability to build relationships in order to scale.
00:20:34
Speaker
These are the things that will determine our value as human beings in the sense of business. I think that conversation is just becoming reality now.
00:20:54
Speaker
Yeah, I, this very unfortunate for kind of the last 15 to 20 years worth of college grads, I feel like it might've just been starting when I graduated.
00:21:07
Speaker
but this neoliberal fever dream that there should be an ah ROI to learning, right? Like go in, plug into this skill, be this highly competent cog in the grand machine and see a return on the, that is the wrong bargain. And it's a category error as to why you learn things, right?
00:21:30
Speaker
And so if you think about, the day-to-day going from analysis, if machines take that over, to orchestration between machines. Orchestrators need to be able to ask great questions, think deeply about problems or challenges or situations.
00:21:47
Speaker
The purpose of higher education is to engage with challenging material to learn better things. And it doesn't fit... Nicely into the neoliberal economic frame of the individual and the realized self through ah roi blah, blah, blah.
00:22:03
Speaker
And it was a lie. They were sold a bill of goods. Right. And ah i I hate that for them. But I do think there is now this reckoning and colleges have ah have a, are their hands
Impact of Remote Work on Human Connections
00:22:17
Speaker
are not clean, right? They played into the business economics of it. They tried to tout all the degrees that they were doing and they would say like, our graduates get this much job placement, blah, blah, blah.
00:22:27
Speaker
They played into that narrative and now they're like, oh, right, we need to teach people that career is a a journey and you need to be equipped for that journey rather than this end point that you will just reach because you have a piece of paper in your hands.
00:22:40
Speaker
It pisses me off to no end. Well, yeah. And I think, again, you know, part of part of what's missing in the whole experience, like what we were maybe looking for, again, I appreciate being a remote worker. I do appreciate like having the balance of my life and not having to transit to and from work every day.
00:22:59
Speaker
There's the hybrid balance, right? Where like at least once every few weeks I go meet up with my team or would be fine occasion to do so, like budget planning. Um, um There is something to be missed though about learning to deal with people every single day in a place. yes
Importance of Genuine Human Connections
00:23:15
Speaker
We are enclosed together and you got to get things done and you might not like one another, but you got to work together and you got to get over yourselves and you got to figure it out just enough to get it done.
00:23:27
Speaker
Yeah, I do think the remote work is a luxury that should be afforded later in your career, right? Like, i appreciated earlier in my actual real work career, like, having to go to a place and, like, interact with those people, for sure. You know, and, you know, funny enough, there was a time, i remember when i I got my first director job, it was, like, my honest-to-goodness envisioned dream that I'd, like, made of reality for a short period of time where I lived... about two and a half kilometers walking distance to my office downtown in Ottawa. I had this nice office. It was big, whatever, a nice fancy title.
00:24:03
Speaker
I got to walk to work and walk home and the weather was great. And if it was winter time, I got to skate on the canal and because that's the thing you can do in Ottawa, you can skate to and from work off the That's sick. That's so fun. That's a separate conversation, but it's fun.
00:24:17
Speaker
um but I lived this weird, like fantasy of just like, cool. I'm living this like life that I grew up watching on TV and I thought it was the coolest thing ever. And, you know, I got to wear my fancy little suit, you know, and say good morning to the secretaries and all that kind of thing. Um, yeah.
00:24:34
Speaker
And you realize that like, I, sometime I'd say COVID broke it. It shattered the dreams of everyone. I still feel like we're living in a fever dream. That's March, 2020. And I kind of can't wait to wake up, but I think, I think we have to redefine expectations. And as a human society, we have not known how to do that because the profiteering of AI has driven us to such a point that we,
00:24:59
Speaker
we have allowed ourselves to lose track of what um the human experience is. Like not only in a professional sense, but in like an actual connected sense where like people don't know how to just go to an event or be in a place and see someone that they might be interested in talking to and just being like, Hey man, how's it going? Or Hey, how are you? Yeah. Just literally start a conversation. Start a conversation. And not not actually have any like purpose of a thing you're trying to get out of it. like I'll tell you an interesting story. my um One of my best friends back home in Kingston, guys worked in a bar his whole career.
00:25:35
Speaker
He's a bar guy. He's great. He's beloved though in the town. He's great at it. Super good athlete and stuff too. Doesn't know a thing about tech. Doesn't do anything about tech. Most of my like actual best friends outside of work have nothing to do with tech thankfully.
00:25:47
Speaker
And so there's another guy his that used to be his housemate that moved over to Calgary. He started a really cool business. He built um an environmentally friendly um ah proprietary um ah portable energy machine or power generation machine. Really Yeah, yeah.
00:26:03
Speaker
And so that guy's in town in Ottawa for the CanSec conference. And he's just like, hey, man, one of my one of my other best friends, he's go to be in town. Are you going in Ottawa? Right. And this is like the day before.
00:26:15
Speaker
you going to be in Ottawa for this? I'm like, whatever. I fly back from Toronto. Yeah, I'll have some time. and he's like, do you mind just like meeting up with the guy? Like, I just think it'd be good if you guys got together. Like he knows that I do something in tech. He knows his other friend does something in tech. His friend's a mechanical engineer. Mutual interest of a sort. Just something in common between on two people that he really, really liked. He was like, do you mind just making time to meet up with them?
00:26:38
Speaker
Right. And Kenny's like never really asked me to do this. And so i was like, all right, well, sure. the other Any friend of yours is a friend of mine. And so I've made time yesterday and we sat down had a coffee. And it was a really good time.
00:26:50
Speaker
And for half an hour, I sat with this complete stranger who's in a completely different industry than us. And the mutual experience that we had was we're friends, a lot the same friends. Because, you know, he went to university in my hometown.
00:27:03
Speaker
And we just, they became friends. And it's like, cool, maybe we can help each other out. Maybe not. But otherwise, like, you're just a cool dude. And like, let's just stay in touch. If I'm in Calgary, I'll come see you. If you're in Ottawa, let's hang out.
00:27:15
Speaker
Yeah. and wait Way to hit the themes of the last ah the last episode that you weren't part of. but that's like But that's like exactly what David Homan was talking about, is this ability to connect dots between people and connect at a very human level.
00:27:29
Speaker
But I think, George, I think like part of this whole... thing with AI and and where we keep telling people just learn to code and become machines and all this shit. This is gross. We've lost that. Because like really like whether or not end up doing anything to help, you know, my new friend's business or whether not he does anything to help me, it's irrelevant.
00:27:48
Speaker
The point is I got to meet a cool guy that my friend vouched for. So I knew he was going to be cool. And now I have this friend who's in another city, other side of the country. And if I'm in that city, I can go say hi and hang out with
Engineers as Creative Problem-Solvers
00:28:00
Speaker
him. And I got so much value out of that experience.
00:28:02
Speaker
and I don't think people time for that anymore. No. and And because again, there's just sort of like this transactional, like what can you do for me yesterday sort of mindset, which is just very depleting and impoverished.
00:28:15
Speaker
But also what I would, the counter argument to that, like just learn to code plus AI is going to replace software engineers. I don't know how many of our listeners work with software engineers. Yeah.
00:28:28
Speaker
But the best software engineers i have ever worked with can have very deep conversations about the problem that they are trying to solve. Emphasis on the word engineer, right? They are problem solvers. And yes, they are very proficient at code, but like the big brain creative knowledge work part of their job is like whiteboarding and mapping through problems and trying to understand what somebody else in the business is trying to accomplish, which is not something that...
00:28:57
Speaker
is just lines of code, right? That's anyway.
Pope Leo on Human Dignity and AI Ethics
00:29:00
Speaker
um That brings us to the humanities and we would be remiss if we did not end the system update with some kind of reflection on the Pope.
00:29:12
Speaker
We have Magnifica Humanitas, right? It's ah over 40,000 words. It is Pope Leo's and first encyclical and it is about human dignity in the age of AI.
00:29:26
Speaker
If you haven't read it, I encourage you to do so. I'm not a believer. I still thought it was an amazing document. Um, He draws these two parallels, right? He says in ah the book of Genesis, we have the Tower of Babel.
00:29:40
Speaker
Everyone should know these stories, right? Even if you aren't a Christian or a Catholic. But he likens this moment to two choices, right? We can either build the Tower of Babel, right? Disconnected from one another, trying to do this grandiose project to achieve...
00:29:57
Speaker
super intelligence, machine God, whatever they want to call it, some a very vague outcome to touch the divine. or and he invokes um the story from the book of Nehemiah, where as a community, they are given separate tasks. And as a community, they rebuild the walls around Jerusalem.
00:30:22
Speaker
And so it's this tension between concentration of power, a grandiose project or communion. And I don't know, it's in ah it's an amazing document.
00:30:32
Speaker
ah He is to be commended for it. I have some issues with, you know, co-founders Anthropic being there. There's a lot to be said about power, et cetera. Catholic Church obviously, you know, has some stories, but
Technology's Moral Imperatives
00:30:44
Speaker
I don't know. want to get your take on it. um It seems like a very important cultural moment for sure.
00:30:50
Speaker
Yeah, and like i I am a believer. I'm a Catholic, so um I'm a very big fan of Pope Leo, and I love whatever gets produced of Chicago, um so thank you for that. um and And shout out to him. like This doesn't come as a surprise to him, right? So I will comment ah from a position of faith for one second, which I rarely get to do on any kind of anything work-related, so this is kind of fun for a sec. but this is Yeah, man.
00:31:16
Speaker
This is a Pope who...
00:31:20
Speaker
goes out and washes the feet of prostitutes. but That's a real thing. And you have to understand the significance of that. Like, just like how John Paul used to forgive and visit prisoners in prison, right? Like this is- Including the man who took a shot at him.
00:31:36
Speaker
Including the man who took a shot at him, right? So we have to understand the the characters that we're dealing with. And in this world of nothing but retribution and revenge and spite, this is truly refreshing.
00:31:49
Speaker
I'll say this, like my thought on this poem. As to what you just put out in terms of the humanitas, the Vatican took the AI question more seriously than most boards have.
00:32:02
Speaker
Or governments. Or governments. Like 42,000 words, five chapters signed on the 135th anniversary of, I believe it was the Rerum Nivorem to the day.
00:32:16
Speaker
And that's that's that's that's not a hot take. that's that's in That's institutional intent from the church. A thousand percent. Leo showed up to present it himself, which the Vatican notes is uncustomary. Popes usually delegate this to cardinals.
00:32:31
Speaker
ah He's putting his personal weight on the document, which tells you how seriously he wants it read. um I believe Chris Ola from Anthropics spoke at the presentation and called for was it informed critics who will tell the labs when we are failing.
00:32:49
Speaker
We need moral voices ah that the incentives cannot bend. right That's an AI Lab co-founder publicly asking for external moral guardrails.
00:33:01
Speaker
I think people need to to go back in this episode, hear that again. AI Lab co-founder is asking for morality to guardrail what they produce.
00:33:14
Speaker
Yes. i mean, some of that feels a little like abdication, but I take your point that there is a serious... effort to at least couch the conversation and something other than like efficiency and compute.
00:33:30
Speaker
right i would I would say like the core argument is like technology is never neutral. It it yeah takes on the character of those who build, fund, and deploy it. right So it's a line that that should make every founder, every C-suite, and especially technical C-suite, sit up.
00:33:46
Speaker
The Vatican is calling out is calling up that like that the tech is not just a tool. yes to To call it such is a cop-out that the industry has- Yeah, it's a cultural artifact.
00:33:58
Speaker
you they've they've been hiding behind this for 30 years and the tool argument died on May 25th. You know, it maps directly to, to kind of what you and I do on separate other projects, you know, when it comes to, to, you know, training data provenance, you know, if the tech, if the tech carries the character who built it, then, you know, knowing who built it and what went into it isn't a compliance checkbox. That's a moral baseline.
00:34:24
Speaker
And that has to be considered, you the deeper play, um, in the encyclical is is naming the danger that humans start seeing each other as products.
00:34:36
Speaker
And that's not a religious framing. That's the same argument Tristan Harris has been making for 10 years, you know, just for better Latin.
Balancing Technology and Human Values
00:34:44
Speaker
Yeah, it's, it's, it's, he is trying to speak to,
00:34:50
Speaker
The zeitgeist that has us all feeling whether but we are believers or not that like something in our life is missing, right? Something we are bereft of meaning, right?
00:35:03
Speaker
ah We've talked about hustle culture. We've talked about in this case, the productization of attention, and right? are we What are we? What is the thing? And if, if you will indulge me, George, I do have a few things that I want to read out of it because I went to town on this and highlighted it. Please, sir. um You did the homework.
00:35:23
Speaker
Yes. So ah from section 99, this is a quote, but I will also sort of skip around. We must avoid the misconception of equating this type of intelligence with that of human beings.
00:35:36
Speaker
These systems merely imitate certain functions of human intelligence. In doing so, they often surpass human intelligence in speed and computational capacity. Yet, this power remains entirely tied to data processing. So-called artificial intelligences do not undergo experiences, do not possess a body, do not feel joy or pain, do not mature through relationships, as you said, and do not know from within what love, work, friendship, or responsibility mean.
00:36:08
Speaker
Bars, Pope Leo. But also, let me... let me I'm going to keep going here. ah
00:36:22
Speaker
The speed and ease with which answers or summaries can be obtained risk extinguishing the desire to ask questions, which is a process that bears fruit only over time.
00:36:34
Speaker
As Plato wrote, the deepest and most important things are learned only after much time and effort by engaging in discussion with others, striking upon ideas and experiences together like flint until the spark of understanding is kindled within us.
00:36:47
Speaker
We must learn then how to exercise restraint in the use of AI and to protect our young people from the promise of the perfect machine, from that subtle temptation which renders human thought seemingly superfluous precisely when it is most needed." end quote I mean, there's a lot in here. It's kind of amazing.
00:37:10
Speaker
I'm here for it. I'm here for it. um But yeah. So I think we'll we'll end the episode by saying we've covered a lot of ground here, but I think all four pieces ah are saying the same thing from different vantage points, right? We have this sudden vibe shift. It's attached to some hard economic truths.
00:37:31
Speaker
And in examining those economic truths, I think we are forced to rethink what is the role of human beings in the economy vis-a-vis technology. And then we have this encyclical here that is really a call to arms in some way of returning to a life devoted to establishing meaning rather than just economic output. In other words, i think we are, we're currently sort of building something we can't measure. We can't really justify it. And now we're kind of like scrambling to define what the humans are for.
00:38:04
Speaker
Yeah, I'd say that's kind of the the biggest point and should probably be on a bumper sticker. Like I'd land hard on this. Like the industry skipped the what is this for question and went straight to, you know, how big can we make it?
00:38:19
Speaker
And so now the bill is due and nobody has an answer. So the, the, You know, what humans are for piece is the real reckoning. We outsource that question to philosophers and then defunded the philosophy departments. And now we're asking VCs and prompt engineers to answer it. And like, it's not going to end well.
00:38:36
Speaker
And so I'd say the honest move for any, any leader right now isn't the biggest side. It's to admit we're in the um the early innings of a 30-year question.
00:38:47
Speaker
And, you know, it's it's it's the operators who are willing to sit with the ambiguity instead of selling certainty. They're the ones are going to be standing 2030. And that's the truth. Yeah, man.
00:38:59
Speaker
Yeah, man. um So to our listeners, if you got kids, if you're young kids, how do you prepare? i have really unhelpful advice. You have to be an interesting person. My great hope is that what this does is it sort of rekindles Renaissance education. If the future is about asking the right questions of these machines and orchestrating things,
00:39:18
Speaker
You have to know a lot about a lot. You can't just be single threaded. So read widely, ask a lot of questions, encourage your kids to ask a lot of questions. I know it's exhausting when they keep asking why, but it's an important skill.
00:39:31
Speaker
And um just keep... the focus on the humans. As we have said time and again, this is a tech podcast about humans. Technology is pretty awesome. Every Wednesday, I've been trying to highlight what AI can actually do that materially impacts lives rather than just like chatbot girlfriends.
00:39:47
Speaker
So ah keep interested in technology, but don't take it at face value. Go touch grass, read books, live a good life.
Podcast Conclusion and Call to Action
00:39:56
Speaker
All right.
00:39:57
Speaker
And that with that, the old men will get off their soapboxes and we will see you again next week. If you like this conversation, share it with friends and subscribe wherever you get your podcasts for a weekly ballistic payload of snark, insights, and laughs.
00:40:14
Speaker
New episodes of Bare Knuckles and Brass Tacks drop every Monday. If you're already subscribed, thank you for your support and your swagger. Please consider leaving a rating or a review. it helps others find the show.
00:40:27
Speaker
We'll catch you next week, but until then, stay real.