Future of workforce: How tech’s bets on AI will impact labor

Future of workforce: How tech’s bets on AI will impact labor

AI-Generated Summary

Microsoft’s recent layoffs, impacting nearly 4% of its workforce, reflect a strategic shift toward AI, though experts debate whether AI is the direct cause. CNBC contributors suggest these cuts are part of broader corporate restructuring during technological transformations, similar to past shifts like cloud computing. While AI is enhancing productivity, it’s not yet replacing jobs at scale. Software engineers, for instance, spend minimal time coding, focusing instead on testing and collaboration. The discussion also touches on future AI integration, such as AI “bosses,” but cultural and technical hurdles remain. For now, AI serves as a productivity tool rather than a job eliminator, with more restructuring likely ahead.

📜 Full Transcript

>> Welcome back to Squawk Box. This week Microsoft announcing it is laying off nearly 4% of its workforce as it tries to refocus more on AI. CNBC contributor Mikael Liberum is a Silicon Valley based journalist. She joins us live this morning. We’ve also got CNBC’s Steve Kovach here at the Nasdaq. Is this the beginning? We got the jobs number later. So that actually sort of plays into this whole situation. Do we believe that AI is the reason? I think it’s the excuse maybe. >> Not a reason. I look at like what Marc Benioff said the other day and he said 50% of our work right now is being done by artificial intelligence. If that was true, we’d be hearing about massive layoffs at Salesforce. We’d be hearing about huge improvements to margins and so forth. But we’re hearing it increasingly. The Journal did a story on this yesterday, guys. They said, you know, saying the quiet part out loud. Ford CEO is talking about it now. We had Andy Jassy from Amazon talking about it. Those Microsoft layoffs though yesterday. No word of AI. They did talk about efficiencies and things like that, leveraging technology presumably AI, but it was mostly just getting rid of middle managers. I can’t replace that. >> Michael, what do you think? >> So I agree, but I don’t think the only options are AI or BS. You know, I think there’s a truth somewhere in the middle too. So while I don’t believe that these are job losses due to AI, that AI is taking over jobs and yes, totally agree that we would see a bloodbath if that was really starting to happen. I do think it’s the result of the company reorganizing itself around AI, and we saw the same phenomenon in other technological transformations. 2014. You know, Microsoft shed 18,000 jobs as they were making the transition to really focus on mobile and cloud. And so this is kind of a natural thing that happens with Silicon Valley companies when we’re going through one of these big technological transformations and reallocations. >> But if we were having this conversation 12 months from now, do you think we’re going to see a lot more of this? And will it be much more severe? >> I think so. I don’t know about more severe, but I think, you know, Microsoft has now gone through several waves. You know, maybe there’ll be more, maybe not. By the way, the cuts are really across the board, both geographically and across different divisions. Definitely hearing about some cuts on the Xbox side, various gaming studios. So they’re they’re across the board. I do think we’ll see more of them. But you know, that said, again, I don’t know a single software engineer out here who’s actually lost their job to AI. And even if 20 to 30% of the code is being generated at Microsoft now by AI, you know, software engineers do a lot of things. They’re not just constantly writing code nonstop. I asked a friend of mine just the other day who’s a software engineer out here, how much of your time are you actually like, you know, giving off to coding? And she said, about 10%, you know, they’re checking the software, they’re testing it. They do a lot of other things. So I do think that there’s going to be a lot more, you know, in other companies as well. I don’t know about Microsoft, more of this restructuring and reprioritization, but not a huge loss of jobs due to AI anytime soon. >> It seems like it would be a good time for a lot of the other maj7 names, or big cap tech names outside of meta to examine that year of efficiency. Meta really gained a lot in terms of stock market value because of the year of efficiency. And with AI kicking in, the ability to reap those efficiencies much easier. I would imagine. >> We hear about that all the time, but we just don’t see the evidence of it yet. I mean, there’s so many anecdotal stories when I talk to CTOs at companies and I say, you know, are you guys using, are you using, did you buy a copilot license and so forth. And so many people say, yeah, we’re testing it, but it’s too expensive. It doesn’t do enough. You know, we see some moderate gains in productivity. It’s not really fulfilling that promise that we heard of. But let’s extrapolate this a little bit because I was talking to I can’t really say who it was, but I was talking to someone who works in HR at one of these big tech companies. They are already talking about a universe not in the not too distant future, where your boss is an AI agent and they’re already talking about within this company planning for a world where you’re literally reporting to an artificial intelligence agent, what does that look like? How do you handle human resources? How do you manage people like that? This is how they’re already these companies are already thinking that way. >> I haven’t talked to people about that. I’ve talked to people about who who think that they’re going to have sort of these agents hierarchies where there’ll be an AI boss that would be bossing around, if you will, other AI agents that would have different tasks. If you were to think of sort of a various hierarchy one might be responsible for if you were a reporter, one might be a reporter, and they might have an editor or a producer, or in this. >> Why wouldn’t you make an AI agent. >> That does it all? Yeah. Why can’t it do it all? >> Why did it have. Why is there many be layers of agents and bureaucracy? >> I think early on. Part of it is, as you probably, we’ve all learned through prompting, you have to do it is everything is actually quite specialized. You can’t if you just tell the tell AI to go do something. >> You have to be specific. >> It doesn’t it can’t really do it right. It’s just like a person. By the way, a person is specialized in something and they have a they have sort of a strategy and a plan to do it if you. It’s not like it becomes an all seeing robot, I don’t think. >> And to your point, that’s why we’re not seeing huge layoffs because it can’t take those jobs yet. >> Mikael, what do you think about that? Do you think we’re all going to be have have an AI boss soon? >> No, I don’t, not at all. I think we will have an AI, you know, helper, an AI assistant. I think that makes a lot more sense to me. By the way, there’s a company out here called Lattice Works. They. They’re in the air space, and they actually launched a product a while back that was onboarding AI agents. And there was such a backlash. They decided to pull the product. They’re heavy into AI and using AI for all sorts of HR, you know, practices, functions. But that one didn’t go over well. I think as a culture, you know, corporate cultures, even on the startup level, like not quite ready for that, but but yes, we’ll have more and more of AI, a lot of experimentation going on right now, not a lot of use cases that are really proving themselves quite yet. But I do think we’ll continue to see people working hand in hand. Vibe coding is real. I mean, engineers are leaning on it and I think it’s going to get better and better and it’s going to have more and more use cases. But I absolutely don’t think that we’re going to be, you know, getting