Anthropic CEO Warns AI Could Wipe Out Entry-Level Jobs

Additional Resources

• Dario Amodei’s The Adolescence of Technology Essay
• BLS’ State JOLTS
• Haver Analytics Article

Transcript

Pete Newsome: 0:00

Welcome to Cornering the Job Market for Tuesday, January 27th. I’m Pete Newsome, and there is a big story that just came out today. It’s from Dario Amodei, the CEO of Anthropic. Now they’re an AI powerhouse. If you aren’t familiar with Anthropic, they’re probably the hottest company in that entire space right now, due to Claude Code, which is just taking over development, as well as Co-Work, a new product that sounds like it does pretty much everything for you. I haven’t used it yet, but it is filling up my Twitter feed lately. So Dario wrote an essay. It’s not the first one he’s written, but this thing was a monster. It’s 36 pages, about 20,000 words, I think. I read it first thing this morning, and it’s called The Adolescence of Technology, and it essentially serves as a dire warning uh about the perils of AI. His framing on this is important. He’s not making the usual argument that technology changes jobs. I mean, that’s a story we’ve lived, most of us have, right? Through the internet, you know, through computers. Technology has changed jobs in human lives for many, many years, right? Generations. Technology is a lot more sophisticated now. But his argument is that this current AI, this superintelligence, is going to be different because of the speed of development that we can’t keep up with as humans. It’s breadth, and you know, basically who’s going to benefit and who will lose. And essentially, a very select few will benefit from this. Most will will not.

1:41

Amodei says AI will drive just rapid and dramatic economic change. I mean, the paper was about a lot more than just the job market. But what I thought was interesting, uh I thought the whole thing was fascinating, quite frankly, but on displacement specifically, he reiterates a prediction that got a lot of attention about a year ago, where he said AI could displace half of all entry-level white-collar jobs within one to five years. And his concern, and this goes way, way beyond job loss, right? It’s what it represents for humanity as a whole. And I will tell you that when I hear people, as happens constantly now in discussions about AI, get tend to get hung up on the time frame. So when you say one to five years and then one year passes, people say, well, see, that didn’t happen, right? This is much ado about nothing. Oh, AI is overhyped. But this is such a blip of time and the big scheme of things. I mean, my youngest child of four is a senior in high school. He’s going to major in biomedical engineering, he’s a lot smarter than I am. And he’s going to be in school five years from now. He’s probably going to be in school eight years from now. So whether it’s one year or five, is there really a difference? No, there’s just not. And you either buy into what he’s saying or you don’t. I personally do. I talk about that a lot. And this is something that I highly recommend listening to. I mean, I could talk about this all day. Like I said, it was 30, it’s 36 pages, but check it out. Darioamade.com. Uh you can find it. It’s well worth listening to. And listen, I this is a guy who, you know, it’s not in his interest to talk about how dangerous AI is. I mean, this is his business.

3:33

And meanwhile, we have other people. I mention this a lot too. David Sachs, the AI czar, or what I typically think of as the White House’s resident AI hype man, uh, goes on these media tours talking about how it’s AI is not going to take jobs, it’s going to create jobs. And one of the points that Amade makes that that’s something that I say often, I think about a lot of in terms of AI, is yeah, there’s going to be a need for new jobs, but don’t you think AI is going to do better at those two? And that’s really his point. I mean, he says that this is going to be 10 to 100, if not more times faster than humans at everything. And he compares it to having, you know, a country of 50 million people all at the top of their game, right? The Nobel Prize winners and statesmen and just the smartest, all the minds of the planet, all together at once, in one AI system. So if you’ve seen the show Pluribus, boy, this really starts, it really starts to come to mind uh as you read this. So that is the biggest headline of the day by far. This paper, I guarantee, is going to get a lot, a lot of attention in the weeks and months to come. I mean, it’s it’s something all of us need to read, in my opinion. Uh, but also this uh Bureau of Labor Statistics put out the state job openings um and unemployment numbers. Nothing really happened with unemployment level month over month. What I did find particularly interesting, what jumped out at me from this report, is a massive um drop in job openings in Florida. Great, where I live, don’t like seeing that. 73,000 jobs um were lost month over month.

5:19

I mean, that is definitely the headline of the BLS data that came out this morning. So, look, slower hiring, fewer people getting interviews, longer time to fill, bad for the job market, bad for someone like me who owns a staffing company. Um not a great start, but this is this is November data, right? So the problem with these reports are we don’t get them, we get them so late, right? We’re already in January, so I expect that Nov uh December is probably gonna be more of the same. Now we get the uh the weekly numbers from ADP. Uh those just came out too, and you know it’s it’s it’s more of the same, right? They now have the numbers through January 3rd. So basically the last week of December shows that there were just under 8,000 jobs uh added per week for the moving four-week average. So again, there’s not I I really there’s not much takeaway from that, right? Yes, people stop hiring around Thanksgiving and Christmas. Those numbers aren’t nearly as interesting to me as what comes next, right? What comes at the beginning of January. So we’re gonna see that uh as the reports start to come out over the next couple weeks, and I’m very eager to see how that looks. Um so that’s so that’s kind of it today for the headlines. I mean, there was a um a report that came out uh from uh Haver Analytics uh where the author argued that if you want to understand where the job market is heading, then you should look at real wages and aggregate hours work.

6:52

Do you look at those things combined? And what the report showed was that wage and hour, or I’m sorry, wage and salary personal personal income growth slowed to 3.8% year over year by November, which is down 5.5% from the year before. So that’s a significant drop. 5.5%. That’s a big number. You usually don’t see that year over year. Um and he said with inflation at 2.7%, real disposable income growth from wages slowed to just 1.1%. So we’re not keeping up. Um at the same time, job growth weakened significantly in the second half of 2025. So I’ve been talking about that a lot lately. All the numbers say that. We really took a big dip. Um the government shut down, got I mean, just created a mess, and I don’t think we really recovered uh last year. So look, the the the administration likes to say how great things are going, but it’s just not in the job data from last year. I know that inflation seems to be under control, and there’s a lot of positive signs. The stock market is is up, precious metals are up, the price of gas is down. I mean, these are all great things, right? We haven’t really seen high inflation, but the jobs numbers in particular just seem to be going nowhere. Um, and that that’s something that I personally care about as much, if not more, than any of these other numbers. So nothing really good to report. I’m glad 2025’s over. Let’s get into the new year. Is everyone good for that? Great, let’s do it because here we are.

8:19

It’s almost the end of January. It’s crazy that we got here. But those are the headlines for today. Thank you for listening, as always. Before we wrap up, here is a fun fact for the day: the post-it note. The post-it note exists because of a failure. It started as an attempt to create a super strong adhesive and instead resulted in one that was reusable and just barely sticky. I I feel like post-it notes are always barely sticky, right? Mine don’t, mine don’t stick. They’re always falling off when I try to use them. But um, hey, sometimes a miss turns out to be the product, and that was it for the post-it note, and that is it for today. So thank you again for listening to Coring the Job Market. Please like and follow me. Subscribe if you know anyone who’s potentially interested, please share it with that too. Them too. Look forward to talking to you soon.