I don't think you can treat owners of the same shares differently in the way this is suggesting. The VC shareholders and the employee shareholders are probably on equal footing and getting the same price. VCs will own preferred but I doubt that is enough to windfall them at the expense of the common shareholders.<p>So if VCs are getting paid a certain share price, employees with vested stock almost certainly are getting the same price. And probably employees with vested options can either exercise now or will just get paid the net during the transaction.<p>Yes, the company is probably doomed so people staying there are not doing well, but they also just got paid a 3x premium on their vested equity.
Yes I think you are right here. The purchase price is high enough for all parties to be get return on their shares, and whilst there will be a waterfall for who gets paid first, I doubt many people will be unhappy with this deal.<p>Unlike Windsurf... who's 2nd employee only got 1% of what their shares were worth (<a href="https://news.ycombinator.com/item?id=44673296">https://news.ycombinator.com/item?id=44673296</a>)
i thought so at first, but I did some digging and changed my mind. it's possible the following is how it goes:<p>- secondary transaction with the preferred shareholders (VCs) at some price that implies a 20b valuation<p>- founders quit and get new employment agreements<p>- some cash is transferred to the company as a license fee<p>- no acquisition means no DOJ approval<p>in this scenario the headline can be $20b but the cash expense can be much lower, you have full flexibility to direct whatever cash or equity you want to founders vs the rest of the company, as an up front payment or as retention/salary, and the founders have no hinderance from working on anything they touched at previous company because of IP license.<p>I actually bet this is how it went down. This is becoming the standard in the industry and it's just awful for the future of SV
Don’t the founders (and the board) still have fiduciary duty to the common holders?<p>You can’t stop the founders from leaving, but selling the crown jewel IP in a transaction that doesn’t benefit the shareholders seems a stretch.
> fiduciary duty to the common holders?<p>as long as the transaction is reasonable, they've held up this fiduciary duty.<p>And the minority holders will need to sue for damages in any case, it's not an "automatic" crime. The cost of that suit will be more than the value of the gains and damages awarded.<p>Therefore, minority shareholders in a startup are highly likely to get screwed - not to mention they don't get a say in decisions being made at the top.<p>The only thing preventing this is social pressure (ala, reputational damage, if the founder did it). And if the payday is high enough, the reputational damage is irrelevant (you'd be out of the game with a big enough payday!)
To make this right they’d just have to amend the first part to “secondary transaction with shareholders at some price that implies a 20b valuation”.<p>Has there been any evidence yet that the VCs got paid for their shares but the left behind employees didn’t?
Wouldn't this imply that the founder's don't get paid either? The acquirer would simply need to have buy-in from the investors to make the deal happen, and the founder would need an offer that is bigger than any other possible "soft landing."
Doesn’t this depend on how the ip was structured? If it was kept as a separate entity, or the firm named ownership of the ip in nonstandard terms, then they could pay investors but not employees.<p>Unfortunately, we could likely find thousands of different ways not to pay employees given they don’t have board seats, and are typically on non standard equity.
It’s also possible that both are correct, and the deal is actually illegal. It’s pretty common for deals to push close to the line to extract maximum value for one set of parties, and sometimes this is misjudged.<p>I guess we just need to wait and see if the common holders are happy or sue.
EXCELLENT analysis Ossama<p>>"Non-exclusive" means no monopoly concerns (anyone can license Groq's tech)<p>- except that you can bet only Nvidia gets the absolute top of the line architechture and design - - - - - all others get 2nd best or worse.<p>>The "non-exclusive" label is legal fiction. When you acquire all the IP and hire everyone who knows >how to use it, exclusivity doesn't matter.<p>But the “non exclusive” part is what significantly weakens any case the US DOJ may consider bringing forth, if at all..<p>If I was in the Nvidia camp I would be admiring how brillant the strategy was all formulated, in fact, I have to believe that IP attorney's were consulted on how best to avoid DOJ scrutiny.<p>On the other hand, there will be those who can see how this limits competition. It would be interesting to have some of our HN attorneys weigh on on this deal.<p>As you said about the remaining employees: . . . Their equity is worthless. . . <they> got nothing while Chamath made $2B. Is Chamath a conniving scoundrel ? I'll let others judge. Maybe someday we'll see Zuckerberg and Chamath in the ring together - - Elon seems to have bowed out.
I wonder how the startup scene will adjust to this if it becomes mainstream. can employee contracts be modified to force compensation even in this case? seems difficult to write one up without weird second order effects.<p>if this does end up being something that is legal and successfully circumvents anti trust, does it mean antitrust actually is a failure in practice?<p>2026 hasn't even begun and more shenanigans are in flight.
This won't last. It's just creative regulatory evasion and the loopholes will get closed.
Regulations shouldn’t have to change with every new fintech innovation, and the IRS already has the necessary laws in place to catch and prosecute abuse of unpredicted loophole technicalities as tax fraud with intent. We’re better off applying a general tax to “gross revenue not paid as wages” and directing it to a universal basic income fund. The purchase price is gross revenue, and the less they pay out as employee wages, the more the (not subject to deductions) tax charge becomes. Sure, they might still fuck over workers, but at least the workers could afford to quit (thanks, UBI) — and VCs would face the choice of taking half the payment and fucking over workers, or taking the same size payment while <i>not</i> fucking over workers. They may be selfish, but they’re not so self-destructive as to choose the former out of spite: it would utterly destroy their ability to hire brilliance in the future, especially once people can afford to say no.
> Regulations shouldn’t have to change with every new fintech innovation…<p>Hard disagree given that a lot of fintech innovation is increasingly devious ways to circumvent the spirit and the letter of the law<p>> … and VCs would face the choice of taking half the payment and fucking over workers, or taking the same size payment while not fucking over workers. They may be selfish, but there not so self-destructive as to choose the former out of spite<p>Also hard disagree. The VC and investor people I’ve met and work with seemed to have a cultural aversion to labor being anywhere near the same level of compensation or power as them. I would fully expect them to take a deal that fucked over the employees if they got paid the same either way.<p>You’d have to tune your suggested system so that not fucking over the employees was heavily incentivized
Works for me! I do tend to be more optimistic than most. I would scale it so that the tax scale is applied with an exponential factor that concentrates the impact on high-revenue businesses while protecting low-revenue ones. (And I suppose as a bonus it would provide a financial counter-incentive against merged conglomerates, too.)
[flagged]
This comment is not cool, and is not welcome on HN.<p>You had a history of guidelines-breaking comments and moderator warnings up until a few years ago, and we've not seen any comments from you until the past month or so, and now you're back into those bad old patterns. You are of course welcome to participate here. But this is only a place where people want to participate because we have clear guidelines, and most people take care to observe them. Please do your part to raise standards here rather than dragging them downward.
if we use your logic then every law written would be failure since at some point people would discover loophole/flaw that would get abused<p>the fix is we use more ambiguous words or just stronger government control<p>see how China "control" capitalist on this case if you want absolute government control
Mr Kwok already analysed these deals and gave them a nice little acronym: HALO<p><a href="https://kwokchain.com/2025/07/15/the-halo-effect/" rel="nofollow">https://kwokchain.com/2025/07/15/the-halo-effect/</a>
Related:<p><i>Nvidia to buy assets from Groq for $20B cash</i> - <a href="https://news.ycombinator.com/item?id=46379183">https://news.ycombinator.com/item?id=46379183</a> - Dec 2025 (400 comments)<p><i>Nvidia just paid $20B for a company that missed its revenue target by 75%</i> - <a href="https://news.ycombinator.com/item?id=46403041">https://news.ycombinator.com/item?id=46403041</a> - Dec 2025 (133 comments)
IANAL, and am especially weak on US law, but I suspect this is only an antitrust loophole if the administration chooses not to act. Substance over form must apply? Pretty sure this wouldn't fly in European law.
Sure, but the US has a history of not acting.<p>If it was a normal acquisition, it would automatically trigger anti-trust investigations. Under the current administration, I think it is unlikely the acquisition would be blocked (although it probably should be...), but it would involve more bureaucracy, and would take longer.
So many companies doing non-"acquisitions" during this AI boom!
Though this one is at least more comprehensive than say, Google simply hiring back Noam Shazeer from Character.AI or OpenAI taking Windsurf
Google also hired the core Windsurf engineering and research team, not OpenAI:<p><a href="https://www.reuters.com/business/google-hires-windsurf-ceo-researchers-advance-ai-ambitions-2025-07-11/" rel="nofollow">https://www.reuters.com/business/google-hires-windsurf-ceo-r...</a>
What is keeping Google/Amazon/Microsoft from licensing Groq’s tech? Sans people to be sure but at substantially reduced price. The Groq Cloud people should owe no allegiance to Ross.<p>BTW, the FA says Nvidia bought the patents. That’s probably an overstatement. Grow said non-exclusive licensing and Nvidia hasn’t said anything.<p>I think Nvidia licensed the IP and ‘bought’ (handcuffed) the people.
> Groq built the region's largest inference cluster in eight days in December 2024. From that Dammam facility, GroqCloud serves "nearly four billion people regionally adjacent to the KSA." This isn't API access. This is critical AI infrastructure for a nation-state, funded by the Public Investment Fund, processing inference workloads at national scale.<p>Maybe I'm just completely out of touch, and hardware has never been my expertise, but does it take O(days) and not O(years) to build data centers these days? I know Grok DCs in Memphis were built under a year cutting many corners and using plenty loopholes, but even by those standards, bringing up a full data center in just over a week sounds impossible without some insane construction automation to me.
Physically building a datacenter yes. Repurposing a crypto datacenter for a new purpose, no. But days is hyperbolic no matter how you look at it unless they’re reusing ALL the existing infrastructure (network/security/hvac/physical server racks/some amount of compute). And aren’t actually including the time to procure the inferencing chips.<p>You can call up Cisco tomorrow and offer them a billion dollars to get you an entire datacenter worth of switches tomorrow and the answer is going to be no because they just don’t keep that much inventory sitting around. That’s why covid was such a shitshow.<p>I should give a caveat of: they could in theory redirect existing orders to you but would likely be violating contractual obligations and risk a lawsuit from the fortune 500 Peter they robbed to pay you Mr. Paul.
"Smart" accounting by not counting design, planning, procurement timelines would explain it, that makes sense.<p>If that's the case, it's saddening to see details like this hyped up to borderline fraudulent levels.
they are likely referring to their hardware deployment. Which <i>can</i> be done in days. The datacenter is owned and operated by someone else and they just bring in the racks.
This behavior is extremely damaging to the startup scene. Who would join a startup these days unless it’s run by a close friend or relative? At least in that case, the scorned junior employees would have social recourse.
It's not like larger companies don't also screw over their employees in various ways. After having to leave AWS due to my fully distributed team that was formed during WFH being forced to "return" to an office that most of ever never lived near, I've preferred working for smaller companies not because I care about equity (I'm in the fortunate position that I can survive comfortably and save for retirement on my salary rather than needing to rely on the value from options/RSUs), but because my confidence in my ability to predict where things are headed goes down increasingly with each additional level of management between me and whoever has the power to arbitrarily decide to upend my employment on a whim. In the long run, I'll probably be fine if my employer doesn't make me rich, so as long as my projected retirement age isn't actively getting pushed back based on my current income and spending, I'd rather optimize for minimizing the likelihood I suddenly find myself unemployed due to untenable working conditions or getting unexpectedly laid off. My experience at startups has been that it's a lot easier to tell when things might start to get dicey several months down the line and start to prepare for if I need to find another gig. With a large company, I've seen that happen much more suddenly for people who had no reason to suspect they might need to in advance.
I don't think this makes it much worse because that's hard to do, it's already terrible. Getting screwed by startup founders has been the status quo for at least 15 or 20 years now.<p>If you're just a worker then demand fair market wages, work healthy hours, and treat your useless class of shares as already used and discarded scratch off lottery tickets.
If you join a startup, and have equity that isn’t special in some way (defending against liquidation preference or dilution), you’re the sucker. You’re just going to grind for someone else’s payday when a deal is made in a room you’re not in. You’ll only be made rich if someone with the power to drive the decision thinks you should be. As always, it’s who you know and being likable.
I’ve thought about this comment, and am replying to it to amend it (as the edit window has passed). I made this comment based on observations I’ve seen during this AI investment bubble and before it. Most times, common shareholders get the short end of the stick. But I will add, there are some “less than the majority” situations where I have seen employee shareholders treated with dignity and respect, and provided access to liquidity accordingly, and I would be remiss if I did not call that out. It will be challenging to know ahead of time, but there are decent people out there who won’t take advantage or use their power against you economically (imho). “Be lucky” is unfortunately not actionable.
Right? As a former founder, I laugh every time I get a 'Founding Engineer' recruit email...
It is definitely time to stop looking at equity as part of pay at a startup. The trend is extremely clear, startups aren't paying out to employees but the C suite gets internal raises and IPO is pushed to infinity. It is nice to have some paper laying around but that is all it is, paper. Go to a startup for a year. Get the experience, move and get a 20-50% pay increase and keep doing that every year and you will be way happier and financially healthier.
This is always the case. Negotiate equity, but assume it’s worth zero. It’s not liquid and highly speculative. It’s a nice to have.<p>edit: which doesn’t mean join companies you don’t believe in! Please do. But don’t expect it to be there, don’t include it in life plans, don’t pay attention to valuations, etc.
I thought this was always the case? Hearing about examples certainly isn't new.
It has always been the case, but each year there’s a fresh crop of new, bright-eyed 20-year-olds who haven’t learned it yet. The entire startup ecosystem essentially depends on the fact that some people haven’t yet internalized that options are worthless and working 80+-hour weeks if you’re employee #3 or higher never pays off, because even in the slim chance your company has a successful exit you’ll get fucked over by antics like this.<p>The best we can do is try and make “options have an EV of 0, startups aren’t worth it, join a FAANG” a widely-known meme in places like HN to keep as many people as possible from having to lean this the hard way. We’ll never save everyone, but at least it’s more widely-known than it used to be.
It hasn't actually always been the case and the real issue is the false advertising that you actually have equity. If my equity of 1% was real then I would get value as the company grew but the reality is that options/shares without some sort of exit is worth 0. Founders and the C suite often (always now?) get 'internal' raises meaning when a new round of funding hits they get to sell but nobody else does. This, to me, completely destroys the concept that equity is an incentive to build the company and means it should -never- be used as part of a hiring pitch since the people pitching it, founders and the upper management, obviously don't believe in it themselves. If you really want to see if the leadership believes in the junk they are telling you then ask them to put in writing that internal raises are available to all at the same percentages or they are available to nobody.
Getting screwed over is a part of life.<p>The problem imo is when you actively lie to me. I won't go into specifics, but I was lied to , said F em, went the legal route and got smacked down.<p>It's not even worth a name and shame. Just sip a shot of whiskey and try to move on. This is why I like contract jobs. Ain't no equity. It's much more honest.
Yeah unless you are a founder or top investor it’s pretty much a guarantee that there will be no exit.
If you need a job for things like food and housing a startup is cool.<p>I fully expect to be lied to repeatedly though about my own pay, our prospects, etc. I had to learn the hard way that these lies are defacto legal because employees won't realistically be able to sue.<p>But hey, the base pay is probably enough.
I was talking with a great-sounding few-person early startup (nice people, non-evil business, interesting work, etc.), and they wanted me to fill a highly-skilled role... in-office in a VHCOLA, for $110K and "0.5%" in usual option schedule. (Presumably also with the usual barriers to options ever being exercised or liquidated equitably.)<p>Even fresh grads with no experience take home more in this town.<p>I live to work, and I'd be willing to spend a few more years in student-apartment quality of life, and to work like a strategic asset to make the startup successful. But I've learned that deal should include a FIRE lottery ticket, not a condo downpayment lottery ticket.<p>If your early startup doesn't want to share significant equity, <a href="https://levels.fyi/" rel="nofollow">https://levels.fyi/</a> provides TC numbers of what established companies are paying, even for people who wouldn't be good for a startup.<p>Maybe it's the recent years of what VC culture has devolved to. ("Why is your cap table cutting in early key hires significantly? Do you have a leadership problem, bro?") Maybe this is just another facet of the "mask-off" or "late-stage capitalism" that people have started calling out in other facets of society.
Oh FFS 0.1% of this acquisition is $20M. 0.5% is $100M. Junior to senior equity lies in this range. They'll be more than fine. They'll be 1%ers to 0.1%ers after taxes, yeesh. It's never ever enough. is it?
Those numbers are not realistic. At a company at this stage (series E I think?), you’ll be lucky to have 0.01% as an engineer.<p>Most importantly, there is no guarantee there will be any payout at all. It’s not an acquisition and we don’t know the terms.
Series E was just 3 months ago. $2M for 3 months work seems fantastic to me. Series B equity was anytime through early 2021. This is a fantastic outcome for everyone in Groq.<p>What I'm reading here is 100% envy and resentment of their success. But that only works if you're already rich or president, preferably both for best results.
> GroqCloud will wind down over 12-18 months. They'll either get laid off or jump ship to wherever they can land. They built the LPU architecture, contributed to the compiler stack, supported the infrastructure, and got nothing while Chamath made $2B.<p>This is depressing.
Unlike other SaaS "acquisitions" of late, this will be not as straightforward closure of a subscription business.<p>The $1.5B contract with the Saudi Arabia is substantial and investors will want that monetized too, there are also existing DCs GroqCloud have in the ME region and also other spots around the world that are quite valuable for their hardware and power agreements etc.<p>Nvidia has CIFUS and other regulatory concerns and also don't want to compete against their customers be a neo cloud provider, the Saudis likely still want their DC build outs to proceed.<p>All this to say, the remaining parts while no longer as glamorous is still worth a lot and cannot be easily sold to big tech co. GroqCloud is more be like Nokia Technologies/ Networks rather be killed.<p>---<p>As a result, staff not part of the Nvidia deal likely have solid jobs and also now the opportunity to climb the ladder quickly now that a lot of leadership positions have opened up.<p>They are also going to have to be compensated higher in cash or poached by an upcoming chip startup as they are no longer tied to equity options vesting scheduled of a very valuable company (Pre deal Groq or now Nvidia).<p>In any scenario they will come out better of the deal, not as much they could have in a full acquisition yes, but certainly better than most engineers not working for a hyped AI startup nonetheless.
That’s just it.<p>We’ve entered a new era. Big companies don’t need your startup. They only need your smart guys. Just those few guys. You keep the rest of your engineers and figure out what to do with them.<p>And lately, the answer has been, “wind it all down”.<p>This sucks so bad for most of their employees. But it’s a signal to the labor market:<p>Be very honest about what you are when you’re considering working at an AI startup. Are you an AI expert? Or a TF/Pytorch monkey? There’s an enormous difference between those two things. If you’re not the key guy, require a good salary up front. Because I don’t see a future where the “acquiring” companies start needing you as well.
> But it’s a signal to the labor market:<p>Or... Maybe we should start to think about how we let corporations get bigger and bigger? What happens if an entity (read: company) becomes so valuable, that it is basically indestructible? Does it have the power to change politics to their discretion? And as such, also influence the legislative?<p>I find that highly concerning.
Nvidia didn't buy Groq because they need any "smart guys" at all.
I think it's important to note that there's nothing forbidding LPU style determinism from being used in training. They just didn't make that choice.<p>Also tenstorrent could be a viable challenger in this space. It seems to me that their NoC and their chips could be mostly deterministic as long as you don't start adding in branches
You're right but my understanding is that Groq's LPU architecture makes it inference-only in practice.<p>Like Groq's chips only have 230MB of SRAM per chip vs 80GB on an H100, training is memory hungry as you need to hold model weights + gradients + optimizer states + intermediate activations.
Would SRAM make weight updates prohibitive vs DRAM?
Read the article and where it talks about accelerated vesting of Groq shares for both the leadership team that goes to Nvidia and the regular employees that stay at Groq. Is that even guaranteed? It's not an IPO or an acquisition, so why would vesting schedules change?
I would assume there's no accelerated vesting but there's also nothing stopping Nvidia from issuing refreshers of equal value to unvested options. That's been a common recruiting tactic for a long time.
Very interesting would love to read about this, do you have a link to the article?
> The "non-exclusive" label is legal fiction. When you acquire all the IP and hire everyone who knows how to use it, exclusivity doesn't matter.<p>I have some doubts about this point. IP is IP, independent of the people who invented it. If a different hardware company were to also pay for a non-exclusive IP license, maybe it will just take a few months to catch up. It’s like inheriting a codebase written by another team, and there will be some pain and some time needed to integrate it.<p>In fact if GroqCloud wishes to survive, it should very well just attract licensees for its IP and collect license fees for the foreseeable future.
A la carte in AI is going to be the name of the game for a couple reasons:<p>- Avoids regulatory scrutiny (for now at least)<p>- Nobody is actually entrenched enough for customers to matter<p>- Weird "celebrity" culture in tech, and AI especially. Everyone is looking for a "whisperer" or a "godfather" or whatever.<p>- Investors still get paid out<p>Smart operational talent will probably adapt by demanding higher salary, signing bonuses, severance packages in lieu of equity. Distribution of the true "lottery tickets" will get more uneven.
I'd like to call back to yesterday's discussion on IP law [1] sparked by recent comments from Rob Pike.<p>There was a major thread on the issue of regulatory regimes and the dysfunction that can arise. How is this acquisition not a textbook example of said dysfunction? This non-acquisition acquisition does not happen at all in a world without IP law.<p>I think we're seeing a culmination of the dysfunction that results from IP law. The sheer amount of capital has given unbelievable momentum to the forces of consolidation. I still can't foresee the endgame (who can?) but it's even harder to see how it'll turn out well.<p>[1] <a href="https://news.ycombinator.com/item?id=46396075">https://news.ycombinator.com/item?id=46396075</a>
So... What will the actual impact on groq services be?<p>I'm a fan, and I use Groq a lot for systems I build. I think they offer something different to most other providers (cheaper, faster, and until recently "we don't store your data by default") and it will be sad to see that fade.
This has nothing to do with antitrust. Not like the current administration is going to enforce it anyways. Nvidia simply wants Groq’s tech and leadership without the burden of 500+ employees.
Feels like a frenzy!
To me, the willingness to spend 20bn for a different architecture speaks to the competition in the market, across the big players, which doesn’t seem to be factored into their valuations.
I wonder how much of the cap table knew<p>Matthew Berman (youtuber) mentioned he's invested in groq and found out same time as everyone else. Guessing he's a small/indirect investor but still telling
So I guess this talent hire + tech license is the new way to acquire startups? Chatacter.ai, Windsurf, and now Groq.<p>Any employees of those companies lurking here? I'm curious how the morale is now.
Nvidia bought groq? I am out of the loop but what does groq have that they want?
Nvidia hired all their top personnel and paid $20B to license their technology, but stopped short of actually acquiring the company. Very similar to how Google didn't buy Windsurf.<p>I assume it's more about what Groq had that Nvidia <i>didn't</i> want, which was competition (in inference hardware).
I'll just make note here for anyone else confused that Groq and Grok are distinct entities. They just have similar names.<p>Groq is more of a hardware focused company.
A chip dedicated to inferencing that's better than their GPUs.
Nvidia losses 200+ billions on recent weeks because Google TPU actually good and market realize that Nvidia have no "moat"<p>this is just panic buying to make stronger foothold
Also worth thinking a about the private equity market scene, groq was afaik tradable be it thinn liquidity on platforms like equityzen. What did those shareholders get?
>Timing it for Christmas Eve ensures minimal media scrutiny of these connections.<p>Sounds like the media is truly the one in charge.
No shade but most other coverage will focus on whether this signals an AI bubble. That's missing the story.<p>Nvidia explicitly did NOT acquire Groq. They licensed the IP and hired the talent. This structure dodges CFIUS review (Groq had $1.5B in Saudi government contracts), antitrust scrutiny, and years of regulatory delays.<p>The $13B premium over the September valuation was the cost of regulatory arbitrage. Announced Christmas Eve while Trump's AI Czar (Chamath's All-In podcast co-host) is in office. Chamath's Social Capital made AT LEAST ~$2B on this exit.<p>My article breaks down: what Nvidia actually bought vs what they left behind, why the deal structure matters, who got paid, and the political connections nobody's talking about.
It will be interesting to see if this is an exit for investors and which ones. Given it wasn’t an acquisition but licensing.
> Why do I hate Chamath?<p>>Let's look at the sh he dumped on retail with his abysmal SPAC track record<p>I do not see why one would feel animosity towards Chamath for this reason. Was there fraud involved? Otherwise, all investors are liable for doing their own due diligence.
Nobody in finance holds a high opinion of Chamath.
Part of it is the irrational feeling of a disgruntled investor, doubt that I'm alone.<p>The other part is he has a track record of dumping on retail then telling them not to buy his next deal once he's already cashed out.
You can feel animosity towards someone without thinking they've met the elements of a crime.
Do you need someone to commit a crime to hate them?<p>Do you still need them to commit a crime when they are rich and powerful, and connected with the current administration that sets the law?<p>If a new law came out that said Chamath can do whatever he wants and it’s always legal, is it impossible to hate him in your worldview?
nVidia's big problem right now is they have more money than they can productively spend and are way down the stupid money hole just looking for any kind of return. This is a failure of capitalism.
The part I don’t fully understand is: could this non-acquisition eventually make the deal less than ideal for Nvidia?<p>Is it really a given that GroqCloud is going to be sunset and the company will die?<p>Couldn’t this company hire talent and continue to operate and maybe even innovate? Couldn’t Groq even hire back some employees from Nvidia? If any of them live in California there’s nothing stopping them and they have a bunch of cash from Nvidia. There are all kinds of loopholes for that like contracting arrangements.<p>Nvidia doesn’t really have exclusive access to any part of the company. They didn’t necessarily remove a competitor, though I’ll grant that they <i>likely</i> did in practice.<p>It’s potentially possible that the regulations did their job and kept a competitor on the market, though again I imagine this is my naevity speaking and that the most likely outcome is that Groq will wither.<p>I also don’t fully understand if the Saudis are getting cashed out or not. Are they really going to roll over and allow their Saudi AI data center to become worthless? I would think they have a lot of motivation after this deal to make sure Groq still operates and serves their goals.
I agree, TFA seems to make a lot of assumptions. I kinda think they are correct just because they support the narrative that everyone involved is carefully taking care of their own, but it sure seems like it could go other ways.
Related:<p><i>Nvidia to buy assets from Groq for $20B cash</i><p><a href="https://news.ycombinator.com/item?id=46379183">https://news.ycombinator.com/item?id=46379183</a>
> Most production AI applications aren't running 405B models. They're running 7B-70B models that need low latency and high throughput.<p>Really? At least for LLMs, most actual usage is concentrated on huge SOTA models. 1 trillion parameters or more. And LLMs seem to be the lion's share of AI compute demand.
[dead]
[flagged]
[dead]
[flagged]
This is about Groq (the semiconductor company), not Grok (xAI’s LLM).
I don't think Musk has anything to do with Groq AI (his Grok model is from xAI)
This is a about Groq, not Grok (Musk's chatbot).
Grok != Groq
The chatgpt-ism and the gpt section headlines make this piece immediately unreadable. Why do you outsource your own blog’s thoughts to a machine? Terrible.