Why AI researchers are getting paid like NBA All-Stars

0:01Hello and welcome to Decoder. This is Alex Heath, your Thursday episode guest host and deputy editor at The Verge. Today, I'm joined by Hayden Field, The Verge's new senior AI reporter. We're going to talk about the AI talent wars and why some researchers are suddenly getting traded like their NBA superstars. If you've been paying attention at all to the AI industry this past month, you've likely seen stories about the hiring frenzy that's been happening between big tech, research labs, and AI startups. Leading the charge here has been Mark Zuckerberg, who has been making some truly unprecedented job offers to try and lure top AI talent to meta.

0:38But there's a whole lot more to this story. Hayden and I have been reporting on this for the past several weeks to get a sense of how big some of these offers really are, why companies like Google are opting to hire instead of acquire, and what it means that some of the most sought after AI experts in the world are no longer motivated by money alone.

1:01Hayden, so good to have you on the show. Your first Decoder appearance. Welcome. Thanks. I'm so excited and honored, especially to be talking about this. It's just crazy out there right now. It is crazy out there for sure. Before we jump into this, Hayden, since you've only been at The Verge for, gosh, what is it, a couple months now? Talk a little bit about your beat. What are you focusing on and what are you excited about? What should people be reaching out to you about? I've been on the AI beat for about five years now. It's been insane to see how much it's changed in that time because I started before ChatGPT was even a thing.

1:31I think the most interesting part of the beat to me is the power dynamics involved and the societal implications. It's crazy how much money is at play here in the hands of so few people leading the charge. And the fact that this is one of the only industries now where you can get a ton of funding, a ton of buy-in without even having a product yet. So I would say, you know, interesting things about bias and systems, ways they're affecting our everyday lives right now that maybe the public should know about, and just crazy stuff about the power dynamics involved, whether it's the executives and how they handle their power and the way that they advance these systems forward with or without the opinions of the people building them, their employees.

2:18and also just kind of the societal implications and how this is starting to affect everyone's everyday life when a few years ago it maybe wasn't. We've both been talking to sources around the AI industry for the last month or so about this talent war that's going on, these crazy offers that researchers are getting. People have seen this mostly as it pertains to Meta and the, you know, 100 plus million dollar offers that Mark Zuckerberg's been making. but there's actually a lot more going on underneath the surface. And it gets to that power dynamic that you just talked about. The amount of money is eye-popping and hard to make sense of if you're not in the industry.

2:57So we're going to talk about all of these key things. I think I would like to start though with Meta, just because it is the way that people have seen the talent wars manifested in the last month or so, what Zuckerberg has been doing. I think it's important really to step back and think about like, why is Meta doing this specifically? And why are they being so aggressive with going out and trying to hire people and spending amounts that surpass what the highest paid CEOs on the S &P 500 make? And I mean, it's obvious if you follow this, that Meta is behind in AI. Llama 3 was a good model, was well-received.

3:39And then Lama 4 earlier this year was not. And actually internally, what I heard was that Lama 4 in the user's eyes was worse than Lama 3. And it actually hurt the engagement for meta AI, which is Zuckerberg's would-be chat GPT killer. And considering that everything is AI these days, and that the stock value of these trillion dollar companies is tied to their AI initiatives, That's not good. And if you're behind on the model race, you're behind on potentially winning AGI, super intelligence, whatever you want to call it. And Zuck was obviously not wanting that to happen. So he's going out and canvassing and has basically hit up every hot AI startup and respected AI researcher you could think of, finally assembled a team.

4:29And yeah, I would be curious, Hayden, when you saw these headlines first coming out about meta, what was your reaction? So I think that's a great point about Zuck wanting to avoid being embarrassed. I think he already put so much into the metaverse stuff. He was like, I can't lose on AI too. But when I first saw these headlines coming out, I was thinking I wasn't surprised in some ways because there's so much money flowing in the industry right now, whether you're pre-profit, pre-product, pre-everything right now, that it didn't surprise me that some of the biggest movers and shakers, the biggest builders in this space are getting recruited this hard.

5:07What I was interested in is how many people are really getting these crazy offers versus how many are just getting normal offers. Normal for the AI industry is still obviously really high, but nowhere near these crazy numbers that we saw from a select few. So yeah, when I did some digging, I saw all is not really what it seems here. Only the real top brass, the top AI researchers that have built the viral products that we all know have been getting these crazy offers. AI researchers specifically, not just engineers, it's specifically researchers are mostly getting these crazy offers. Yeah, as you brought up, buyouts and performance metrics are a big part of this too.

5:48There's a lot of unseen kind of caveats in these offers. So it depends on your performance, how long you stay. And some of them are pretty subjective. So I think that's an interesting, you know, kind of side part of the story here. Subjective, you mean in terms of the performance aspects of the offers? Right. Some of them are tied to specific performance metrics that may be a little bit vague or may be decided upon by the people above you. That's what I'm hearing, at least from people that have turned down these offers. They weren't fully sure on what it would really entail. And so, you know, they would rather just stay where they know what's what.

6:22It's important to distinguish the researchers from the engineers, the people who are researching frontier models. There's not a lot of them. This is actually a supply and demand issue. Based on my conversations, it's easy to say that there's probably less than 2 ,000 people in the world helping build a frontier model from the research perspective. So when you have that kind of imbalance of the amount of money that is invested into the space and the people who can actually build it, they become a very, very hot commodity. So that makes these numbers less crazy in the context of the money that's being spent when Meta, Google, Microsoft are spending tens of billions a year on data centers, on data to build these models.

7:07Spending a few billion on the talent to run all of that is not as crazy. The meta offers are interesting, and you talked about this. They're not as just clear cut as people make them out to be. Zuck is not just walking up to a respected AI researcher and giving them a guaranteed half a billion dollar payout over four years. These are mostly performance related stock awards. In many cases, they have aggressive clawbacks where if someone leaves early, they have to forfeit a lot of the money. Even that signing bonus that a lot of people are getting, which is really the thing that I've heard a lot of AI researchers at competing labs are lured by is getting$100 million up front to just join a company.

7:49But it's not really guaranteed. If you leave, you have to pay it back. If you don't hit certain performance metrics, you won't be guaranteed that full payout. So it's more complex than it's made out to be. And I think in that context, the offers and the numbers are not as crazy. They're still crazy, I think, to anyone who's not in the AI world, but they're not as crazy. Right, exactly. And there's also a lot of boomerang and bouncing around. Like, I was talking to an AI consultant, and he knew someone hired for, you know, a couple hundred thousand dollars as a growth executive at an AI firm. So not a researcher, not an engineer, but got flown to an international offsite.

8:28And then he was cut two months later due to lack of sales. So those performance metrics affect everyone, but they obviously uniquely affect those people that are getting those huge signing bonuses. And, you know, it's going to cut you back a lot if you don't deliver. Yeah. And delivering for a company like Meta is realistically adding trillions of dollars in market cap. It's getting to super intelligence, whatever that means. I'm not sure the industry actually knows what that means. But building this kind of all-seeing, all-powerful AI and making Meta the undisputed winner in the consumer AI race, which means when you think of AI and interacting with it, you don't think of ChatGPT, you think of Meta.

9:09That's going to be a very hard goal to crack. Especially with the name Meta. On that note, we have to take a short break. We'll be right back.

9:28Support for this show comes from LinkedIn. Imagine if any of the movies that included the line, I need the right person for the job, settled for, I'll just take about anyone. How many heists would have failed? How many deals would have fallen through? How many secret spy missions would have ended in disaster? So why would you accept just anyone when hiring for your business? When you need the right person for the job, you can turn to LinkedIn Jobs. And now LinkedIn Jobs is stepping things up with their new AI assistant. So you can feel confident you're finding top talent that you can't find anywhere else.

10:01With LinkedIn Jobs AI assistant, you can skip the confusing steps and recruiting jargon. It filters through applicants based on criteria you've set for your role and surfaces only the best matches. So you're not stuck sorting through a mountain of resumes. Hire right the first time. Post your job for free at linkedin.com slash partner, then promote it to use LinkedIn Jobs' new AI system, making it easier and faster to find top candidates. That's linkedin.com slash partner to post your job for free. Terms and conditions apply.

10:44Welcome back. This is Alex Heath talking with The Verge's senior AI reporter Hayden Field about the big talent wars bursting out all over the AI scene. Many of these companies are spending absolutely ridiculous amounts of money to hire top talent outright, but that's not the only way they're poaching folks. Big tech has gotten its way through acquisitions for a long time, but now there's something else happening where companies like Meta, Google, and Amazon aren't buying up whole startups. They're just buying the parts they want. Let's move on to another aspect of the talent wars, which is all of these interesting acquisitions that are happening.

11:23And they're acquisitions in a different way. Recently, you had a great scoop for us here at The Verge about Google sweeping in to steal away a team of a very hot AI startup. Talk about that. We saw OpenAI's deal to buy Windsurf that had been, you know, really reported over months and months suddenly off. And instead, Google swooped in and hired the CEO, the co-founder, and a lot of their R &D employees and just brought them onto the Google DeepMind team to build agents. So they're going to be working on Gemini. Google's not going to have a stake in Windsurf or Control, but it's going to take a non-exclusive license to some of Windsor's Tech.

12:04So it's been pretty crazy. Their head of business became their interim CEO. They had to suddenly name a new president. And a lot of the employees were kind of, you know, shocked by this because they had been pretty close for a long time. It had been a team that had known each other for a while. It was a perfect example of how soap opera-esque these hiring wars are starting to become. You know, when you're at a company and you've been there for a while, maybe you're employee number one, employee number two, you see on Twitter that Most of your company, the talent of your company is being sold to Google and you're just kind of left with a shell.

12:37It's really, you know, just an example of the time we're in right now. Let's unpack that a little more. So Winsurf is, you know, one of the hottest coding startups. They were going to sell to OpenAI. That deal fell apart. And then, like you said, kind of like a soap opera, the team gathers and finds out that Google is actually hiring away the CEO and the core AI engineering team. So 40-ish people, the company is like 250 to 300 people. These people are already leaving. They're already out the door. And there's a new CEO and you all figure it out. And then over that weekend, WindSurf sells the rest of itself to another AI coding startup called Cognition.

13:17This happened in a period of like two days over a weekend. And that shows, I think, how sought after these people are that are building these leading AI startups, that the moment there's a flare set up, that there is a group of people looking for a home, someone will swoop in and immediately find them that home. And in the Google case, this wasn't an acquisition, but I think we should call it what it is. This is Google avoiding antitrust scrutiny. Years ago, they would have just acquired all of Winsurf and not used the parts of the company that they didn't want. Big tech can no longer make these kind of acquisitions for the most part.

13:56And so Google came in and said, we'll take the best people for us. We'll pay you a quote unquote licensing fee, which is really just money that then pays back the shareholders as if it was an acquisition that can also be used to pay back employee stock as if they were getting their company sold. And what's left is, like you said, this shell of a company without its beating heart, so to speak, the people that build the AI models and the AI technology to then scramble and pick up the pieces. This is actually a pattern that's been playing out reverse aqua hires. It started with Inflection and Microsoft.

14:35Microsoft hired away the CEO of Inflection to run its AI group, Mustafa Suleiman. And then Amazon has done it twice with two different AI startups. Google did it with Character.ai. Meta just did it with Scale.ai to bring in their new head of AI, Alexander Wang. And Google just did it with Winsurf. I'm curious, with these reverse acquihires that are happening, Winsurf and Google being the latest example, how does this impact the people on the ground at these startups? The people who either decide to go or don't or get left behind without a choice. How does it affect them and then just the broader ecosystem?

15:13because this is a new thing that's starting to happen in Silicon Valley. It's interesting because, you know, AI engineers, AI researchers, anyone in that industry has a sense of loyalty a lot of the time to where they work. You know, sometimes they'll stay somewhere for their whole career so far, or they've known the founders for a long, long time. And that's part of why they took a job. So yeah, there's definitely a sense of, you know, rejection or sometimes betrayal when things like this happen and you're left at your shell company and the person you got to that company to work for, maybe one of the main reasons you made that decision is suddenly gone.

15:49We saw one engineer who said he was employee number two at Windsurf go viral on X for saying he got an offer from Google that exploded within a day. According to some of the replies, including people from Y Combinator, it may have been a$20 million offer as part of a buyout. You know, he said he forfeited his vested shares. He ended up going to Cognition instead. There was some confusion on whether he accepted the offer from Google or whether he turned it down. But either way, he ended up at Cognition. So, yeah, we are seeing people jump ship when things like this happen to them, when they are left with the Shell company.

16:26We're seeing some of them kind of ride the ship down and say, you know what, I'm going to see what happens. I still care about some of the people here. Scale is going through the same thing, you know, after the big deal we saw. So it's kind of a new thing for these AI industry people to be contending with because it's not something that has happened in their industry before. To my knowledge, I don't think any of these startups that have been reverse aqua hired so far are doing objectively better than they were before. Big Tech came in and stole the founding team. Character AI is a great example.

16:58It was a pretty viral chatbot kind of companion platform created by two early Google senior engineers. And, you know, since those guys went back to Google, you don't really hear about character as much. Scale, which Zuck just hired the CEO and some of the core team from for Meta AI's new lab, just did layoffs. You know, Windsurf had to sell itself like within days from Google sweeping in. And so, yeah, there's this, I think, interesting second order effect where big tech is coming in to these often really well-funded, well-positioned startups that are really red hot. They take the team and then the startup is left to pick up the pieces.

17:42And in many cases, they can't. And again, this is like this is a brazen, like middle finger to regulators. Like scale is a great example. My understanding of the scale deal and why it happened so fast and was structured that way is Zuck wanted Alexander Wang in the office immediately. And he was in the office within a few days of that deal being announced, working on setting up Meadows New Lab. There's an expediency to how these deals work where the acquirer can get the people in its tent ASAP. And because the AI race is so red hot and every day counts for who is going to have the best model, who is going to be at the frontier, that speed really matters.

18:24What's interesting with Google, though, Google feels like it's in a pretty good spot. I think Google's feeling, you know, they can come in and do the reverse aqua hire with Windsurf when they want some talent that's available in the market. But I've heard that their deep mind org is not really getting into bidding wars with Zuck like some of the other labs like OpenAI, where a senior researcher compared it to a home invasion, what Meta was doing. Google is, I think their bench of talent is so deep. and Gemini is considered to be such a good model right now that they feel like they can afford to lose people if need be.

19:04And they don't want to get into these crazy bidding wars. So that's been an interesting thing I've just observed about Google recently. I'm curious, especially, I think you've been reporting more on Anthropic recently, which is another huge nucleus of talent that so far seems to be withstanding Zux poaching more than OpenAI and others. Is that another one that we think is in a relatively good spot from a talent perspective, Anthropic? I think so. They have built their reputation on caring a little bit more about safety. As we've seen, people disagree on how different their mission really is from open AIs.

19:39Over the years, it seems to be getting closer and closer. But either way, people who work there really feel that their voice is heard about safety stuff. Top AI researchers really care about safety, and they really care about how their work is being used. So, you know, whether it's making it up the chain or not, the fact that they feel they can talk about it, their voices are being heard, I think is leading to less turnover. And that's what I've seen. I mean, I was looking at a recent report of employee turnover at AI firms and Anthropics was very, very low compared to open AIs and even deep minds.

20:10So I think they are set up pretty well to keep most of their talent for now. They also reportedly just got a way higher valuation, nearly three times their valuation in March. So, you know, people are making money. They feel like their voice is heard, at least for now. We'll see where it goes. But yeah, it does seem like they're pretty set up for now. Yeah. I mean, if you're a leading AI researcher and you're evaluating a really big offer from Meta, potentially hundreds of millions of dollars, but it's heavily tied to performance, right? And you're thinking about the upside to that offer. What are the odds that Meta adds another trillion in market cap versus Anthropic tripling its valuation in the next 18 months?

20:50it's probably more likely that Anthropic's going to go way up faster because Anthropic's just starting from a much smaller valuation. It's still huge, obviously, but it's small in relation to big tech. Especially when the competition is so steep, like you said. I mean, everything OpenAI does, Anthropic and now even XAI are immediately following. They released a government product, Anthropic and XAI both released a government product. Coding help, Enterprise, those are their biggest moneymakers. And now that market is being like, you know, cut in three instead of them just cornering it. So it's definitely interesting to see.

21:27Yeah, I don't think that they have as much runway as they did before in terms of the runway to growth and tripling that type of valuation. We have to take another short break here. We'll be back in just a minute.

21:46Support for the show comes from Rippling. If you're a business owner, here's the truth. SaaS promised to make work easier, but now the average company is buried by hundreds of apps that silo your teams, slow you down, and simply don't work together. That's not SaaS. That's SAD. Software as a disservice. That's why you need Rippling. Rippling is the unified platform for global HR, payroll, IT, and finance. They've helped millions replace their mess of cobbled-together tools with one system designed to give leaders clarity, speed, and control. By uniting your employees, teams, and departments in one system, Rippling removes the bottlenecks, busywork, and silos your software created.

22:30Automated, perfectly in sync, and seriously simple to use, Rippling gives your company one source of truth for your people, their data, and everything they touch. With Rippling, you can run your entire HR, IT, and finance operations as one. Or pick and choose the products that best fill the gaps in your software stack. And right now, you can get six months free when you go to rippling.com slash decoder. Learn more at r-i-p-p-l-i-n-g dot com slash decoder. That's rippling.com slash decoder for six months free. Terms and conditions apply.

23:13Welcome back. Hayden and I are talking about the hot market for top AI researchers right now. Before the break, we were chatting about some of the incredibly lucrative cash offers folks are getting from the biggest companies. But at a certain point, maybe money really isn't everything. You and I have both talked to researchers who have turned down offers that I think to most people, they would say, what in the world are you thinking? I talked to someone who got a personal reach out from Zuck and didn't respond, which I was like props. What motivates these researchers besides money? I do think more motivated by money than they want to let on.

23:53I think when you get to these kinds of numbers, it's impossible to not be. but at the same time, someone just not responding to a Zuck reach out, I think, says a lot about the industry and the way that these researchers are motivated. What have you learned about that? Something that I've seen is pretty constant over the last few years in AI reporting, but is more evident now than ever, is that a lot of these key AI engineers or researchers are what you might call post money. Like they make so much that making more isn't in and of itself a motivator for them. Yes, they're motivated by money, but after a certain point, you know, it's kind of all fake.

24:36You know, you're making millions. Okay. What's another few million? So they care a lot about how their work is being used, who they're doing it with, who they're doing it for, what their company's vision is in terms of accelerationism, societal impact, safety, risk. A lot of them have differing views on that stuff, but it's important to a lot of them that their company or the leader of the team they work on aligns with them on whatever those views are. So yeah, we're seeing people turn down crazy offers, personal reach outs from execs. I've heard OpenAI CEO Sam Altman always calls at least most technical hires, potential hires with an offer, you know, to personally pitch them.