Someone like Elon being asked for their opinion on tech - I kind of understand - was at least at the coal face of SpaceX and Tesla for a time, seemed to understand the tech and was not terrible when it came to direction.
Zuck I'd get, Bezos, Dario, Sam - but I don't actually get why Reid is always in the conversation - he's never been in front of anything
So, a guy with probably one of the largest incentives in the world to trash xAI is trashing xAI?
The information content of this is rather minimal. Even if everything he says is literally true it's hard to tell through the massive, massive vested interest he has.
And it doesn't help that...
'Hoffman, who is an investor in both Anthropic and OpenAI, pushed back firmly on the narrative that the two companies are in a zero-sum race. “We tend to want to tell these stories as cage matches,” he said, as in two companies enter and only one leaves, but “in fact,” he claimed, “there’s a lot of room for both of them to win incredibly.'
that's clearly a very self-interested gloss on the flip side of the situation. Yes, that's in the possibility space. No, I would not consider "both companies do fantastically for many many years" as a terribly large part of the possibility space. Look to all of the many past instances of industries starting up. It is a very common case that if you take the two early leaders you aren't looking at who is going to be the two biggest companies in 10 or 20 years. It is in fact a common case that neither of those companies are the leaders in 10 or 20 years. The sheer staggering size of the AI training moat at the current time may lock in the possibility that no other business could possibly overtake them... but what if somebody solves that massive training gap? It probably isn't mathematically fundamental; I can't help but observe that humans do not get to their level of capability by pouring the entire Internet through their head several times.
He probably does know a lot of things most of us don't know, but I doubt he's sharing very many of them in this article. This is just trash talk.
I truthfully don’t know the answer, but if I had to guess, his connections and positions provide him with an unusual amount of knowledge and perspective. Another might be that his opinions often are correct in hindsight.
I'm not interested in people's take on SpaceX this early after their IPO, they have an ambitious vision and Elon Musk gets a lot of blind hatred. You don't invest into SpaceX to see returns in a month, you're in it for the next five or more years or you're better off finding a different stock to invest in. To date SpaceX is the top leader in getting things into space for the lowest cost, everyone else pales in comparison.
People don't invest in SpaceX because of space launch capability, that barely counts for their valuation at all.
The valuation of SpaceX is due to AI, namely the revenue they get for renting out their GPUs to companies that actually have AI customers, as their own AI tech has not panned out.
For the large number of companies rolled into SpaceX, they are all failed attempts to grow large enough to justify their valuations, and when a company fails to do that it just gets rolled into the conglomerate as a way of hiding the failure.
Tesla's valuation contrasted with its performance means that Tesla will likely be rolled into whatever latest vehicle of Musk's has the most attention, hiding the failure of Tesla to come anywhere near to its promises.
Other than launching satellites, being able to get things into space for the lowest cost is about as relevant as being able to get things to the bottom of the ocean for the lowest cost.
There's no economy in space. It's never going to be cost effective to send anything back down the gravity well, which means that the only way Musk's plan leads anywhere is if he's able to bootstrap an entirely self contained, self perpetrating economy in space. That's not happening in five years.
Edit: and no, data centres in space are not the answer.
This article encouraged me to look at the investor materials [0].
The 55th slide "key metrics" wording stood out to me:
> AI: "Nameplate Compute Draw" Total number of GPUs installed in the data centers at the end of a period multiplied by the respective all-in power draw, reflecting installed capacity and not actual power consumption or utilization
Close to $15 billion in losses since 2023 and not much clarity on actual usage or impact. TIL the plan of record is AI satellites assembled on the moon.
Reid Hoffman hates Musk. Think what you will of Musk (we all have our opinions), but Hoffman criticizing one of Musks companies is the equivalent of Steve Jobs criticizing Windows. Its a personal quibble and therefore not really news worthy.
This is not a very good comparison because Jobs was well known for very pointed and accurate critique of software, which was one of his super powers at Apple. Bill Gates was known for figuring out how to manage software engineering, but nobody would listen to Gates about that, and in fact the only time I ever saw him critique software, talking about the complete usability failure of Windows and Microsoft's supporting websites, it did not require any sort of deep insight.
Hoffmans critique about which businesses have good promise should be taken seriously, if with a grain of salt.
The key part - Reid being invested in both OpenAI and anthropic should have been higher up in the article. Pretty crucial context to him trash talking XAI.
It was at the very top just now when I looked. That said, the site has so many distracting pop-ups and other interruptions it's hard to see anything there.
Compared to what? All the comments seem to agree, but I curious if people here have actually used Grok.
I rotate between major models frequently. Grok has been up there in accuracy and research for some time, trading places with Gemini IMO. Latest 4.3 release has been solid.
Composer is pretty good and now they own Cursor. Don’t count them out yet.
So.. it’s bad, compared to what? Claude from 2 months ago?
Gemini has been atrocious, in my experience. Not sure if it's the harness or the model, but it hallucinates much, much more than GPT (via ChatGPT) or Claude, and weirdly assumes it can just answer complex, knowledge-heavy domain questions without doing a web search.
The problem is that SpaceX financials supporting the IPO say SpaceX is a major AI company that has a minor side-hustle of making rockets that sell satellite internet.
That's the total picture of SpaceX, right. Does SpaceX as a whole make financial sense? No. Everyone knows the SpaceX "value story": AI means that a company that makes a minus 5 billion per year really makes plus 200 billion per year! IN SPACE! But, uh, about those Space parts? Surely those are cashflow positive ... right? RIGHT?
Well, no.
SpaceX it is the 50th or so rocket company. The previous ones did not fail because they couldn't get rockets working or couldn't improve on the state of the art in rocketry. The ones not supported by nation-states failed because they couldn't get the financials working. To be fair some of them failed because they couldn't get to earth orbit. But that's not the common case. More common: "New rocket type works! We demonstrated it succesfully! No launches ... so no money. We're publishing our work and shutting down. Bye". Irritatingly quite a few of these new rocket companies are theoretically more efficient than SpaceX will ever be. Also irritatingly most of these companies, through financial necessity, demonstrated a working rocket in one try, in contrast to SpaceX.
(my favorite? Aerospike nozzles. Aside from their great "Wiley E. Coyote" potential should launch fail they look absolutely incredible)
Did Space part of SpaceX get the financials working? No. Not even with Starlink (their debt repayments still drag it into the negative). What is their fix for too small a market? Make Spaceship, an even bigger rocket ... for a market that sees no use for the existing Falcon 9 launch capacity ...
Starlink: same. It's not even the 10th satellite internet company. The previous ones all failed, because the market was too small, and had to be bailed out by nation states, famously Iridium. Did Starlink solve the financials? No.
The most irritating bit of this is of course Elon Musk himself. Why did he succeed? Well he keeps mentioning himself and "starting from first principles". As illustrated above: he started from first principles, he failed (private, ie. profitable access to earth orbit? SpaceX doesn't do that), then he got incredible amounts of money somewhere to pour down a black hole (using artificial demand like Starlink) and so everything is still moving. Obviously Elon Musk's achievement is 100% finding this money and 0% practicing science from first principles".
That's also Elon Musk's great redeeming quality. What's his achievement? Convincing, first himself, then humanity, or at least enough humans to get ~300 billion in cash, that Space exploration is worth doing despite the fact that it's unprofitable. The actual technical Space exploration side he ... frankly didn't do particularly well, though well enough that it (eventually) worked. But the result is still fantastic: we're in space far more than before!
The SpaceX S-1 says they’re going to be making $320bn by 2030 on AI services at a profit margin of 74%. That dwarfs all the launch business and even Starlink, which they very optimistically project as well. This is how they supported the IPO valuation.
The SpaceX IPO prospectus states that the company is targeting a TAM of $28.5T, equal to roughly a quarter of the world's gross economic output.
Patrick Boyle said it best. Roughly 1 billion people on the planet make more than $12k annually (folks with "discretionary" income). Divide that TAM of $28.5T by 1B and the every single person needs to give SpaceX ~$28.5K every year forever in order for that figure to make sense. It's more than 3x what the planet spends on food currently.
Whilst nobody can dispute the inordinate success Hoffman has had in building and scaling LinkedIn and his work on Greylock, his associations with the Epstein files, previous spats with Musk, and his warped political views makes me question anything he says.
In this instance, I see it as nothing more than a self-serving and politically motivated diss against Musk, even if the substance of what he says is true.
Renting GPUs doesn't make you an Ai company imo, colocation data center is more accurate. One might expect that line of business to commodify within five years.
Sorry, that's a typo, it's not SpaceX, it's SPAC X - as in, Musk is using SpaceX as a SPAC to absorb other AI companies. Cursor is the first, but will certainly not be the last. So, if it's not an AI company yet, it will be soon. I mean, the humongous total addressable market from their IPO filing has to come from somewhere, and Grok will definitely not cut it...
I wake up every morning amazed that there were enough people foolish enough to buy a company with only $18B in revenue and no profit at basically at the valuation of Microsoft (a company with $300B in revenue and $100B in profit).
"The pitch deck, he revealed, describes Manas as “an AI drug discovery factory for creating monopolies” legally permissible, he notes, because pharmaceutical IP functions as a sanctioned monopoly by design."
> The timing of Hoffman’s remarks is pointed. SpaceX went public on June 12th, with AI central to its IPO narrative. Within days, the company announced it was acquiring Cursor, the AI coding tool. Hoffman’s read: that’s not proof of AI capability, but evidence of its absence. “You could almost think of it as the IAC of AI,” he said, invoking the serial acquisitions roll-up strategy of Barry Diller’s internet-era conglomerate. “Use the market cap to buy AI companies and try to buy your way into relevance.”
Careful. I don’t have anti-Musk bias, but it can both be true that SpaceX will be quite successful in the long run and the stock is still overpriced in the short run.
Elon Musk's companies launch rockets, build the most high-tech electric cars, host the world's most relevant social network, deploy high-speed internet to most of the world, build one of the world's leading LLMs, I could go on...
Teslas aren't the most high-tech these days. They have fallen behind on the hardware side, particularly in charging and batteries. Here's a charging speed comparison:
What a terrible world to live in. Of course hes trying to convince Gen Z of the "opportunities" they have. Opportunities for him and his oligarch friends to make a load of $. Cant wait for this AI bubble to crash and dissolve a bunch of undeserved wealth.
Someone like Elon being asked for their opinion on tech - I kind of understand - was at least at the coal face of SpaceX and Tesla for a time, seemed to understand the tech and was not terrible when it came to direction.
Zuck I'd get, Bezos, Dario, Sam - but I don't actually get why Reid is always in the conversation - he's never been in front of anything