2025-07-03 03:34:28
The most incredible thing about the incredible offers – some might call them "Godfather" offers – that Meta is making to some AI talent, are incredibly not the offers themselves. It's that everyone has thus far turned them down.
Now, that very well may change. I mean, they're reported to be offering $300M over 4 years – with perhaps $100M+ up-front in that first year. It has to change at some point, right? Someone has to take that offer? Right? But the fact that no one has yet – none of the people jumping from OpenAI to Meta's new Superintelligence team reportedly got that offer, but instead smaller ones (which are undoubtedly still incredible) – is fascinating. Again, incredible. And it's a really bad look for Meta. Which is, perhaps, why they feel the need to make such an offer!
2025-07-02 17:59:55
Over the weekend, I tried to square some circles around the "$100M Signing Bonus" headlines surrounding Meta's attempts to lure top AI talent to their new "Superintelligence" team. My main takeaway? While the $100M signing bonuses were likely exaggerated, the overall packages also seemed likely from a company, driven by their founder, that clearly felt an existential need to get back into the AI race. Subsequent reporting by Zoë Schiffer seems to further clarify the extraordinary situation. And it's perhaps closer to the exaggerations side!
As Mark Zuckerberg staffs up Meta’s new superintelligence lab, he’s offering top research talent pay packages of up to $300 million over four years, with more than $100 million in total compensation for the first year, WIRED has learned.
Meta has made at least 10 of these staggeringly high offers to OpenAI staffers. One high ranking researcher was pitched on the role of chief scientist but turned it down, according to multiple sources with direct knowledge of the negotiations. While the pay package includes equity, in the first year the stock vests immediately, sources say.
To be clear, that $100M slug may still not technically be a "signing bonus" – which most would consider to be a lump-sum cash payment due at signing – but depending on how it's actually structured, it could be considered close enough.
The "vests immediately" element is clearly key here. That implies an RSU grant with front-loaded vesting for that first year (or two different RSU grants, I suppose). If it's truly "immediately", the employee would get the stock that presumably they could sell at any time (well, in normal employee trading windows outside of earnings season, etc), unless Meta put some sort of restrictions on the sale of the stock? Presumably, they'd want some sort of claw-back mechanism in place in case the employee were to leave immediately – though maybe not if they just really want to make a statement with these offers! (I mean, there must be a claw-back provision, right? RIGHT?!)
Regardless, if the vesting truly is immediate, the employee would also owe taxes on the amount right away (well, at quarterly tax time in the US given the amounts we're talking). And assuming everyone is in California, that will be just north of 50% going bye-bye right off the bat – so they're probably going to want to sell at least some of that stock as soon as they can. Well, unless Meta is also covering the tax burden. They could, but that seems unlikely as this entire thing is already not an efficient way to do comp. (Meta's finance team must be loving this. Not.)
Anyway, with "more than" $100M due in that first year (is that because of an actual signing bonus in cash, or perhaps just their regular old salary/bonus?), the remaining $200M or so would presumably vest over the four year span, per the wording in the report. So that's roughly $50M a year in equity grants. Which would be more than double what Meta's executive team gets (which, of course, is public given that Meta is a public company). It's also more than what almost all CEOs get for all but the largest companies in the world. And also Warner Bros. Discovery.
Okay, so such an offer is rather incredible. Even more incredible is that Meta made said offer to "at least 10" people at OpenAI. Was Meta really ready to commit $3B in comp to these 10 people or (more likely) were they just hoping to entice some of them with the offer, knowing not all of them would take it? In fact, they undoubtedly started with one such offer and then went down a list.
But even more incredible, apparently none of them took it!
On Monday, Mark Zuckerberg sent a note to Meta staff introducing the new superintelligence team. Alexandr Wang, formerly the CEO of Scale AI, is now Meta’s chief AI officer, Zuckerberg said. He’s joined by Nat Friedman who previously led GitHub. Together, Wang and Friedman will colead an organization Zuckerberg dubbed the Meta Superintelligence Labs. The company did not name a chief scientist or a chief research officer as part of the announcement. Neither Wang nor Friedman are thought of as researchers, at least in the traditional sense. None of the OpenAI staffers who left for Meta received the $300 million offer, according to a source with knowledge of the contracts.
That's right, none of the seven people listed as jumping ship from OpenAI to Meta got what I simply must call the "Godfather" offer, apparently. And actually, it's not a "Godfather" offer because those that got it seemingly did refuse! It's not clear who was getting those offers, but presumably it is the kind of researchers that could fill the "Chief Scientist" or "Chief Research Officer" roles.
In yet another scoop on the matter, Schiffer obtained a Sam Altman memo on the OpenAI side about the defections:
“Meta has gotten a few great people for sure, but on the whole, it is hard to overstate how much they didn't get their top people and had to go quite far down their list; they have been trying to recruit people for a super long time, and I've lost track of how many people from here they've tried to get to be their Chief Scientist,” he wrote. “I am proud of how mission-oriented our industry is as a whole; of course there will always be some mercenaries.”
With that, you can probably guess some of the names they're going after within OpenAI, and it appears they have failed so far in those efforts and instead moved down a rung or two. What might those "second-tier" offers have been? A mere $50M upfront? I'm sure we'll find out soon enough given all the leakage (on both sides) here.
Still, assuming the above is all accurate, it seems pretty clear that Altman was at least being more directionally truthful in his (public) comments on the situation than Meta CTO Andrew "Boz" Bosworth was in his (leaked) internal comments on the situation. Boz tried to portray Altman's statement as deceiving. And while the technicalities around the offers may be a bit different, they're still directionally what Altman said they were at a high level, it seems!
As I wrote over the weekend, Boz clearly took his approach to addressing the offers from the position of trying to snuff out any internal disarray around this newfound compensation discrepancy. And again, I think it's probably fair to believe that the "Godfather" offers were meant for key executives Meta would bring on board, not necessarily the rank and file. But that's also hard to say without knowing what those "second-tier" offers – the ones that were accepted – are...
And that leads to one think I'd like to correct from my writing on the matter. Back to Schiffer's first report:
Andrew Bosworth, chief technology officer at Meta, said that not everyone is getting a $100 million offer during a Q&A with employees last week. “Look, you guys, the market's hot. It's not that hot. Okay? So it's just a lie,” he said. “We have a small number of leadership roles that we're hiring for, and those people do command a premium.” He added that the $100 million is not a sign-on bonus, but “all these different things” and noted OpenAI is countering the offers.
The original transcript left out the "lie" part. As such, I wrote:
There's actually quite a lot said – and not said – in those four simple sentences. First, Boz is implying that what Altman is saying about the Meta comp offers just isn't true. But he specifically doesn't say he's "lying" – because he himself notes that there's some level of truth here with the "he's suggesting that we're doing this for every person" which itself suggests that Meta are making such offers for some people – just being "dishonest" as if to suggest that Altman is taking something with some level of truth to it and fudging it a bit to make it less true.
My bad, Boz actually did say Altman was lying. Very directly! And given this new report about the "Godfather" offer and that it was made to over 10 people, I think it's totally fair for Altman to have said what he said:
"They started making these like giant offers to a lot of people on our team. You know, like $100M signing bonuses, more than that comp per year."
So kudos to Meta on the poaching. But the fact that they got no takers on the "Godfather" offer is seemingly a problem for the company. And the fact that they're being disingenuous with their employee base internally seems more so.
One more thing:
“That’s about how much it would take for me to go work at Meta,” says one OpenAI staffer who spoke with WIRED on the condition of anonymity as they aren’t authorized to speak publicly about the company. Other employees said that they were weighing the money against the potential impact they could have at Meta in comparison to OpenAI. Several believed their impact would be greater at OpenAI.
First, a legit lol at that quote. Ouch. Second, it points to the larger element of all this – yes, beyond even the insanely large comp offers. But that's another post...
Update: That other post...
M.G. Siegler
2025-07-01 19:00:15
Step aside, Traitorous Eight, there's a new group of ship jumpers in Silicon Valley. In his memo announcing his new AI organization, the Meta Superintelligence Labs (MSL), Mark Zuckerberg specifically lists out eleven new Meta employees who have come on board for this effort. Well, technically the list is thirteen if you include the new leads of the group, Alexandr Wang and Nat Friedman. But I'm sorry, I need to reference Ocean's 11 with my headline and image here – also, the original – as it's superior to Ocean's 13.
Ruthlessly, Zuck lists out not just the names of the eleven, but where they're coming from and what they did at those old places of employment. Here they are, per the memo obtained by Jonathan Vanian for CNBC:
By my count, that's:
That's... a lot from OpenAI.1 Now you see why they were sending around internal notes that sound quite shaken and stirred – "someone has broken into our home".2 I can't speak directly to the quality of this team, that's above my pay grade (clearly), but it certainly seems more impactful than the downplaying of the poaching a couple weeks back by Sam Altman. It may or may not be their "best people" – but it's a lot, no question. Money wins, and all that.
Beyond that, the most interesting aspects of the memo are what's not mentioned.
Most notably, Friedman's investing partner, Daniel Gross. This immediately jumped out to me because he would be a massive get by Meta, given not just his AI work and investing through the NFDG fund, but also his current role as um, CEO of Safe Superintelligence. You know, the other "superintelligence" effort, started by Ilya Sutskever as his post-OpenAI project. And the one currently valued at $32B despite having just been started and not having any products in market – in fact, they may never have any products in market, but I digress.
It was wild that Gross was apparently going to jump over to Meta. Especially since reports stated that Zuckerberg had tried to hire Sutskever himself (and presumably tried to acquired SSI), but was rebuffed. Divide and conquer, a common tactic used by the Romans in Gaul! Again, ruthless. But the fact that Gross isn't listed here may suggest one of two things. First, either his deal isn't finalized yet.3 Second, that someone, maybe Sutskever, maybe the investors in SSI, got to him and pulled him back from the brink. I suspect we'll hear more about this very soon.
Also not mentioned in the memo: Yann LeCun. Never one to be quiet about the current state of AI, he's been awfully restrained in the past couple of weeks as all of this has swirled around. On the other hand, while he's been against all the talk about achieving AGI through LLMs, he's apparently on board with ASI. Still, it seems weird not to even mention your Chief AI Scientist when you just hired a Chief AI Officer...
Meta's previous model work with Llama does get one paragraph – seven paragraphs into the memo:
I’m excited about the progress we have planned for Llama 4.1 and 4.2. These models power Meta AI, which is used by more than 1 billion monthly actives across our apps and an increasing number of agents across Meta that help improve our products and technology. We’re committed to continuing to build out these models.
Leading with "I'm excited" probably says all you need to know there. But just in case, the billions now being spent to play catch up after the billions spent on Llama drills home the point: Llama, at least as we previously knew it, is yesterday's AI news within Meta.
In parallel, we’re going to start research on our next generation of models to get to the frontier in the next year or so. I’ve spent the past few months meeting top folks across Meta, other AI labs, and promising startups to put together the founding group for this small talent-dense effort. We’re still forming this group and we’ll ask several people across the AI org to join this lab as well.
"It's better to be a pirate than join the navy." We're gonna need a picture of this group together, for good measure. Ideally in suits. But one well-placed bow-tie and a pirate flag will work too. Fine, I'll accept one in front of the Bellagio fountain.
One more thing: one phrase never mentioned once in the entire memo: "open source". Interesting.
Update July 2, 2025: Some further thoughts and analysis on the comp offers...
M.G. Siegler
M.G. Siegler
M.G. Siegler
M.G. Siegler
M.G. Siegler
M.G. Siegler
M.G. Siegler
1 With a special shout-out to Google as 5 worked there at some point, which continues the narrative that modern AI was very much born at Google but they let aspects slip away.
2 And it doesn't include the three OpenAI employees moving over that were previously announced/reported. So that's... 10 from OpenAI alone.
3 The fact that Zuck states that Friedman will "define his role going forward" – not to mention the sloppiness of the whole "lead" and "partner to lead" element for Wang and Friedman around MSL suggests this was all a bit rushed. And that may lend credence to Gross' deal simply not being done. We'll see.
2025-07-01 05:59:51
This past January, I tempted fate. I wrote a headline in the style of Betteridge's law even though I was pretty certain the answer was, in fact, "yes": Should Apple Switch Off Siri? My argument in that piece was that Apple should shut down their internal version of Siri in order to rebuild her from the ground up and in the meantime, outsource her brain to ChatGPT. As I wrote:
And so I ask again: should Apple switch off Siri? Given the aforementioned OpenAI deal, why not fully outsource Siri to ChatGPT? You'd still want to keep the task-oriented elements – setting timers, etc – with her, but everything that requires anything resembling a search query should be outsourced. Right now, you can force this by asking Siri to ask ChatGPT something, but it's cumbersome, I'm suggesting this be made the default action.
By March, with news of yet another delay of an actual useful Siri from 2026 until 2027, I had see enough. This time, there was no question mark: Apple Should Swap Out Siri with ChatGPT.
And, well, here we are. As Mark Gurman reports for Bloomberg today:
Apple is considering using artificial intelligence technology from Anthropic or OpenAI to power a new version of Siri, sidelining its own in-house models in a potentially blockbuster move aimed at turning around its flailing AI effort.
The iPhone maker has talked with both companies about using their large language models for Siri, according to people familiar with the discussions. It has asked them to train versions of their models that could run on Apple’s cloud infrastructure for testing, said the people, who asked not to be identified discussing private deliberations.
While no decision has been made one way or another yet, and work continues on Apple's own internal models for "LLM Siri", this change could happen as early as next year. So why the change of heart? Well, the change of leadership:
The project to evaluate external models was started by Siri chief Mike Rockwell and software engineering head Craig Federighi. They were given oversight of Siri after the duties were removed from the command of John Giannandrea, the company’s AI chief. He was sidelined in the wake of a tepid response to Apple Intelligence and Siri feature delays.
Rockwell, who previously launched the Vision Pro headset, assumed the Siri engineering role in March. After taking over, he instructed his new group to assess whether Siri would do a better job handling queries using Apple’s AI models or third-party technology, including Claude, ChatGPT and Alphabet Inc.’s Google Gemini.
That must have been a fun test. But interestingly, it was not ChatGPT, but apparently Claude that won this particular bake-off:
After multiple rounds of testing, Rockwell and other executives concluded that Anthropic’s technology is most promising for Siri’s needs, the people said. That led Adrian Perica, the company’s vice president of corporate development, to start discussions with Anthropic about using Claude, the people said.
Beyond any technical hold-ups, there's clearly some worry about the message that switching to third-party services would send internally. While Siri has been a mess for a long time, Apple obviously has great talent on their teams, it has clearly been more of a strategy and leadership issue.
On the technical side, it sounds like the biggest hurdle may be getting these third-party models to run on Apple's own "Private Cloud Compute" servers, so that Apple can maintain some credibility on the data security front while outsourcing this work. And Apple would continue using their own models on local devices (though it sounds like OpenAI pitched Apple on this idea too).
But the biggest hold-up?
While discussing a potential arrangement, Apple and Anthropic have disagreed over preliminary financial terms, according to the people. The AI startup is seeking a multibillion-dollar annual fee that increases sharply each year. The struggle to reach a deal has left Apple contemplating working with OpenAI or others if it moves forward with the third-party plan, they said.
A "multibillion-dollar annual fee" let alone one that "increases sharply each year" is probably a bridge too far for Apple. This is a company whose largest M&A deal ever was $3B (for Beats, just over a decade ago). Can you see Apple paying Anthropic – in which Amazon and Google have huge stakes – more than that each year?
And Anthropic should want this contract, as they've struggled to gain traction with Claude versus ChatGPT. Then again, they're seemingly happy to cede that consumer business and focus on coding, agents, and other elements of AI. So maybe they think such a deal only makes sense if Apple makes it worth their while.
That potentially puts OpenAI right back in the driver seat. Perhaps Apple could strike a deal that sees them invest in the company – for real this time.
And perhaps that would set up an unveil that I dreamed up for WWDC 2025, but instead at WWDC 2026:
Can't you just hear them say something like this on stage at WWDC in a few months?
"The response to ChatGPT integration inside of Apple Intelligence has been amazing. So amazing that we're extending the partnership to bring ChatGPT integration even deeper into Siri with [pause for dramatic effect] SiriGPT."
At first, this could be mainly powered by ChatGPT – maybe Apple would couch it a bit, noting that it was a hybrid of their own models and OpenAI's, again, much like Amazon is doing with Alexa+ – but over time, this could slowly morph. Again, perhaps they start to offer other models behind the scenes and seamlessly route queries (which would also ensure they're future-proofed and not putting all their eggs in one basket). And over time, Apple's own models – which they clearly need at this point, even if distilled from others – take over most of the Apple Intelligence workload. But it all starts with ChatGPT giving Siri that lobotomy.
Boom. Ship it.
We're closer than ever to that happening it seems.
M.G. Siegler
M.G. Siegler
M.G. Siegler
M.G. Siegler
M.G. Siegler
2025-07-01 02:26:47
With all the talk of $100M signing bonuses or offers or both or whatever, we're looking right past the issue of "why". That is, why does Mark Zuckerberg have to offer such incredible amounts of money to convince people to come over to work on Meta's new "Superintelligence" team? You keep hearing that it's the market and it's crazy right now. But that's mainly true because it's Meta who are the ones setting this market!
Again, why?
The high level answer is obvious: they need to play catch up. But for months and months seemingly all we've heard out of Meta is how much their products are "killing it" and they're the leaders in AI. Obviously, that wasn't true. But again, why? Seemingly part of it was a fundamental bet which Zuckerberg made that may have backfired. That is, the bet on open source.
2025-06-30 20:20:40
For a few years now, it has been strange that there's a 'MacBook Pro' and 'MacBook Air' but no actual 'MacBook'. I mean, sure, the 'Air' is effectively the MacBook these days, but ever since the last true MacBook was put out to pasture six years ago, it's been a weird branding quirk. And it may be about to change:
Apple’s current entry-level laptop is the $999 MacBook Air, but analyst Ming-Chi Kuo believes Apple is aiming to launch an even more affordable model soon.
He writes on X that Apple will go into production in late 2025 or early 2026 on a new MacBook model that will be powered by the A18 Pro chip, rather than an M-series processor. This is the same chip used in the iPhone 16 Pro line. The machine may feature colorful casing options, including silver, pink and yellow.
Kuo's track-record is mixed on many of his Apple predictions, but that's usually when he tries to pontificate outside of his tried-and-true supply chain sources (he's particularly bad at trying to guess the dates in which things will be released). But if you read his actual post, the above information is clearly tied to a check in with Everwin Precision, which is a Shenzen-based company "that mainly engages in the development, design, production and sales of electronic components," per Reuters. Per Kuo, they're apparently working on both these new would-be MacBooks and the "Smart Glasses" that Apple is rushing towards releasing. So that's promising.
It's less clear how reliable Kuo's information would be about what is powering such machines. But it seems reasonable enough to think that Apple could use the 'A' series chips found in iPhones, as Mayo points out, they're certainly powerful enough, with similar performance to their 'M' series cousins which have been used in MacBooks since they switched to Apple Silicon (the iPad switches between the two series, with the 'Pro' models now getting the 'M' chips).
One big benefit of using the 'A' series chips: Apple produces them at a much higher quantity thanks to the iPhone (and to a lesser extent, the lower-end iPads). So presumably they carry a lower cost and as such, could lead to price savings for an actual MacBook device. But would it be better than, say, using the last generation of the 'M' series chip? Hard to say...
Kuo believes the new product line would ship 5 to 7 million units in 2026, but he's often wrong on such projections as well. If it really is a significantly cheaper MacBook – say $799 – it could be a huge hit for students, one imagines. Especially if Apple is finally looking to bring some fun back into the line up again with colors, as Kuo suggests:
Expected to enter mass production in late 4Q25 or early 1Q26, with an approximately 13-inch display and powered by the A18 Pro processor. Potential casing colors include silver, blue, pink, and yellow
Silver and blue, of course, are already a part of the Air lineup. Pink and yellow would be sort of strange choices – are we sure he's not just thinking about (or hearing about) rose gold and gold? Again, these have already exists before in the MacBook Air lineup so it's hard to say... Presumably if Apple was going to do colors for the MacBook lineup, they would match what they do for the iMac?
I'm more intrigued by the "approximately 13-inch display" – could this mean a return to the 12-inch form factor of the last MacBook? Myself and others loved that footprint, despite the bad "butterfly" keyboard, and pretty terrible battery life. The keyboards have been fixed and Apple Silicon fixes any battery life concerns so... might it be time for another go with the MacBook?
M.G. Siegler
M.G. Siegler
M.G. Siegler
