Traster

(a) that phase out is going to take an extra-ordinarily long time. and (b) it'll be pretty trivial to extend this to charging networks.

Galloway and Swisher have been speculating about this (as well as a Tesla roll up) for a while. It makes tonnes of sense from Musk's perspective- he's got this behemoth monopoly that'll be worth a trillion on the stock market, meanwhile he's got xAI and Twitter which are both sub-scale money losers. It's easiest way to bail out his bad bets (and keep his investors sweet). The only question is whether eventually Tesla joins the party, it makes the most sense for xAI and Tesla to be joined, but I just don't see how that can happen. The benefit of xAI and x.com is that they were private so there weren't meme stock valuations to compete with, the investors would take a resonable payout know, safe in the knowledge SpaceX will IPO to a meme stock valuation. Tesla's valuation has to return to earth before SpaceX can bail it out and by that point it'll be like catching a falling knife.

Almost certainly less than it cost them to train grok and definitely less than they're paying to service X.com's debt load.

Because Elon Musk owns a space company not a shipping company. If Elon Musk happened to own a tonne of boats sure as hell sea-steading data centres would be the future. But he doens't, so it's AAAAIIIII IIIIIN SPPPPPAAAAAAAACE.

I think you're totally wrong on this. Tesla didn't waste the first mover advantage. They benefitted from it whilst it existed, but Electric vehicles turned into a commodity, which was entirely expected and there's no moat.

You've explained yourself why it would be untenable for Musk to pursue becoming the biggest car manufacturer in the world - if he succeeded in that goal... he would have succeded in shrinking the value of the company significantly.

It's pure logic that Tesla has to pursue bets that would justify billion dollar valuations and being a car company isn't that.

I think this article doesn't really engage with why TSMC is in the position it is in. It is in the position it's in due to 2 key reasons. The first is scale, in order to make the R&D worth it you need to sell a hell of a lot of chips. So you need scale in order for the economics to work. One of the things a lot of people point to in the downfall of Intel is they missed mobile and that just cut the legs out of their scale.

The second reason is the opposite though. The industry is cyclical. If you invest a tonne now, and that investment comes to fruition just as you hit the bottom of a cycle you're just going to go bankrupt. You'll have a tonne of capacity, no demand and businesses are designed to go bankrupt in exactly that way. You only need a small surplus of demand to skyrocket prices, and you only need a small deficit of demand to turn into a fire sale.

Let's put it this way - Ben points at a graph and says "Look, here's where ChatGPT got released - and the YoY CapEx went down!". Yes, that is exactly the cyclical nature. It takes a long term to sort the logistics for this industry, so if you see demand in early 2023, that's going to turn into CapEx in 2025 - and it did, a 35% increase in CapEx YoY is yuuuge.

The problem is that if you say competition is the answer what you're really saying is you want one player in the industry to chase a pro-cyclical demand cycle and that will blow them up. The result of this strategy is just to sacrifice the leading players each cycle.

The logical answer is actually, if Google wants TSMC to push CapEx into a bubble, Google should under-write it. Judge where you expect demand to be and hand over cash to buy the chips now. TSMC will build it. If you guarantee revenue in 2029, TSMC will absolutely guarantee you capacity. But Google doesn't beleive it, and they certainly won't pay up front for it. No, what you'll get is special purpose vehicles that look a lot like the big tech companies guaranteeing projects - in a way that doesn't show up on their balance sheets now - but turn out not to be guarantees if it turns sideways. And TSMC can't afford that counter party risk.

It's totally understandable, but it would have been good to have been given a heads up - I'm totally new at the company (new to the industry), assume one of these people knows what they're doing. In the end I got a new boss and convinced him to take a different approach once the first approach failed, but I would not have blamed my new boss for just binning the entire project and getting rid of me.

>I wondered, “Who is this feature even for? Who will use it?” No one on my team knew.

I think there's another key here - Don't assume someone else knows something. If you don't know why something is done some way, find out who does and make sure they do. I've been in so many situations where the organization gets complex - person A is loaned over here or person B is working on project X because team Y needed feature Z. So frequently you'll find out that core assumptions have been made because everyone involved was only half-involved and either kind of assumed someone else was taking care of it, or (more frequently) knows the assumption is wrong but is choosing not to say so for political reasons.

I'm kind of surprised the article doesn't spell out why these are "prediction markets". They're prediction markets because it used to be that what they were doing was illegal. It was illegal to just set up a gambling website in the US and so these "prediction markets" were the same thing as crypto, masking what was really happening behind some tech speak. In the same way that in the crypto market you regularly get behaviour that would get you locked up if it were to occur on the CME, what happens on these prediction markets would be illegal if you were just a betting company.

The reason they're getting big now though, is because we know none of those regulations are going to be enforced. So through regulartory arbitrage, these gambling sites are just going to eat the entire regulated industry until nothing is left.

It used to be Harry Enten's former boss would joke about Scottish teens, when talking about betting markets because frankly, the betting markets didn't seem predictive and were so niche, it seemed like the people on them were generally just betting with no idea what they were betting on.

But today, let's face it, soon we're going to see a contract on whether the S&P goes up or down tomorrow, and when you have that contract, gamblers are going to be betting on it instead of going to Robinhood and buying some Puts.

It's a real test to see if what all those people who talk about the halo effect of trusted well regulated markets are actually correct. The conventional wisdom is that most of the users of these sites are going to get their faces ripped off, and the companies will get very rich doing that.

This seems a bit like a needlessly publicized finding. Surely our baseline assumption is that there are lots of systems that aren't very good at finding cancer. We're interested in findings that are good. You only need 1 good system to adopt. Yes, it's good scientific hygiene to do the study and publish it going "Well, this particular thing isn't good let's move on". But my expectation is you just going until you design a system that does do well and then adopt that system.

If I pluck a guy off the street, get him to analyze a load of MRI scans and he doesn't correctly identify cancer from them I'm not going to publish an article saying "Humans miss X% of breast cancers" am I.

Fundamentally when you think about it, what people know today as AI are things like ChatGPT and all of those products run on cloud infrastructure mainly via the browser or an app. So it makes perfect sense that customers just get confused when you say "This is an AI PC". Like, what a weird thing to say - my smartphone can do ChatGPT why would I buy a PC to do that. It's just a totally confusing selling point. So you ask the question why is it an AI PC and then you have to talk about NPUs, which apart from anything else are confusing (Neural what?) but bring you back to this conversation:

What is an NPU? Oh it's a special bit of hardware to do AI. Oh ok, does it run ChatGPT? Well no, that still happens in the cloud. Ok, so why would I buy this?

And importantly, in this analogy - most people here aren't even able to play that lottery. He founded a company based on the research he did whilst studying for a government funded PhD. Most people are not in a position in their life where they could even spend time trying to do research that would result in this type of eventual wealth.

Firstly,

>Series 3 will be the most broadly adopted and globally available AI PC platform Intel has ever delivered.

The true competitor is Ryzen AI, Nvidia doesn't produce these integrated CPU/GPU/AI products in the PC segment at all.

It is the nature of YC that you're going to get instances like Pickle. YC invests at a very early stage in lots of companies. 40% are literally just an idea. It isn't a scam that one of the companies pivot, it's expected. They're meant to work on their idea, and if it doesn't work or they have a better idea they pivot.

What Pickle is doing is essentially they're falling on the wrong line of "fake it to you make it", it would be totally fine to do what they're doing (allowing pre-orders with a $200 deposit for a Q4 '26 product) if they just weren't lying about the specs. It's pretty clear they aren't going to deliver anything like what they've promised, but that is just ambition. The whole point of YC is that 1 out of 1000 of these companies are going to deliver something revolutionary and you don't get that without 1000 of them trying to do something revolutionary.

Having said that, you only need to watch the launch video to realise the CEO is total moron ("If everyone wore the same pair of glasses, what would they look like?"). But the way YC works, they don't actually have the power to tell Pickle what to do. YC are going to lose their investment on that company.

This stuff is so difficult because it's all situational. In my first job I got hired into a small team, within a year or so my boss quit. I essentially stepped into that job. But in order to be the manager of that team you needed ot be a certain level of seniority, so I was doing the job, but I couldn't have the title or the money. I talked to my manager and made clear that I wanted the job and asked how I get there. The answer was that I needed to be a grade 7 and I was only a (new) grade 5, and since there are only a handful of promotions they can give each year, all things being equal I would get promoted to the seniority level needed to be doing that job in about 6-10 years.

I was essentially doing what this article was advising, but because of the corporate structure I was in, I was just volunteering to be taken advantage of. The correct strategy actually was just to leave. I wasn't going to be successful in that structure, it wasn't a meritocracy and the business results over time went how you would expect. In the end I leverage a job offer to put myself on track to get into that role within 2 years, but in reality that was 2 years wasted, I should've left immediately.

It's pretty clear from other articles the answer is not. Although from what I can see it claims to be about 50 times better than conventional methods. One claim is they tracked a plane to within a few hundreds meters for a 300 mile flight. Which is technically impressive but entirely useless for the stated purpose in this article.

It's kind of annoying that they don't go into specifics about why this is good. The problem with traditional accelerometers is that the error accumulates, and so even small errors accumulate. The article doesn't really address what makes this different - and in fact I don't think it is different. You're still just measuring acceleration using a sensor and that sensor will have errors that accumulate. So the question is how much better is it?

I would wager that actually this is probably just a way of funnelling money into research around quantum rather than genuinely trying to solve this specific problem. This specific problem sounds like it could be solved for a lot less money using conventional accelerometers in combination with some other local location data (optical sensors for example, you're in a very controlled environment).

I was quite surprised the direction this article took. I wasn't expecting reheated whinging about the toolchain.

FPGAs do need a new future. They need a future where someone tapes out an FPGA! Xilinx produced Ultrascale+ over a decade ago and haven't done anything interesting since. Their Versal devices went off a tangent into SoCs, NOCs, AI engines - you know what they didn't do? Build a decent FPGA.

Altera did something ambitious back in 2014 when they proposed the hyper-register design, totally failed to execute on it and have been treading water because of the Intel cluster**. They're now an independent company but literally don't have anyone who knows how to tape out a chip.

I'm less familiar with the Lattice stuff, but since their most advanced product is still 16nm finfet I suspect they aren't doing anything newer than Xilinx or Altera.

We need a company that builds an FPGA. It doesn't matter what tooling you have because the fundamental performance gap between a custom FPGA solution and a CPU or GPU based solution is entirely eaten up by the fact the newest FPGA you can buy is a decade old and inexplicably still tens of thousands of dollars.

If FPGA technology had progressed at the same rate NVidia or Apple had pushed forward CPU/GPU performance, thered be some amazing opportunities to accelerate AI (on the path to then creating ASICs). But they haven't, so all the scaling laws have worked against them and the places they have a performance benefit have shrunk massively.

There's a few things. Let's start with the core of what they say is their value. They have forward deployed engineers - this a totally new, previously unknown innovation - who go to a company, understand their needs and build data processing tools to give them insights. Then, they generalize these tools so that they can essentially sell them as SAAS software, giving them SAAS-type economics.

What other people say is their secret sauce, is they do consulting work for the government (a forward deployed engineer is just a consultant) and they make incredible margins because their senior management and early investors have connections to the government which gets them exclusive access to incredibly juicy contracts. As these contracts paid off they leant heavily into the social media meme stock trend so their CEO spends time talking like a psychopath and doing various non-economic things like spending huge amounts of money running adverts about how they're going to use AI to unleash Americas workers (America's workers aren't able to buy Palantir software or services, but they can buy it's stock).

Ah I see, so the [token holder] hires a [builder] to build something, and uses that to then hire [intellectual] to scam the ['pragmatic user']?

To take this a little more seriously, this is computer programming, very famously you don't need massive gobs of VC capital to build something. The only reason for the [builder] needs [token holder] is to hire [intellectual] to scam [user].

Oh and of course, [token holder] [builder] and [intellectual] are the same guy with 3 different anime profile pics.

There's a difference between a good product and a good business. It's easy to see this is a great product but it's really difficult to see how it's a great business. All the traditional things we look for from great businesses seem to be absent here - they don't have network effects like facebook or uber, they don't have lock-in they way Apple has, and to a large extent they don't have the traditional economics of SAAS. They have a product that is interchangeable with 3 or 4 other companies, they have extremely high initial investment costs to train models, and it's not clear they can actually sell tokens for more than it costs to make them.

It's like the foundry business, an ever increasing cost of moving to the next node requires ever increasing scale which naturally kills off competition.

Training is taking an enormous problem and trying to break it into lots of pieces and managing the data dependency between those pieces. It's solving 1 really hard problem. Inference is the opposite, it's lots of small independent problems. All of this "we have X many widgets connected to Y many high bandwidth optical telescopes" is all a training problem that they need to solve. Inference is "I have 20 tokens and I want to throw them at these 5,000,000 matrix multiplies, oh and I don't care about latency".

I'm actually not that worried about this, because again I would classify this as a problem that already exists. There are already idiots in senior management who pass off bullshit and screw things up. There are natural mechanisms to cope with this, primarily in business reputation - if you're one of those idiots who does this people very quickly start just discounting what you're saying, they might not know how you're wrong, but they learn very quickly to discount what you're saying because they know you can't be trusted to self-check.

I'm not saying that this can't happen and it's not bad. Take a look at nudge theory - the UK government created an entire department and spent enormous amounts of time and money on what they thought was a free lunch - that they could just "nudge" people into doing the things they wanted. So rather than actually solving difficult problems the uk government embarked on decades of pseudo-intellectual self agrandizement. The entire basis of that decades long debacle was based on bullshit data and fake studies. We didn't need AI to fuck it up, we managed it perfectly well by ourselves.

In the UK, there are massive incentives for "tax efficient" purchasing of vehicles. ie, there was a point where everyone who could was driving around in a Porsche Taycan because the tax implications of buying it were comically positive. Of course, those benefits don't translate to the secondary market, so of course there's a glut of 3 year old electric vehicles. No one wants them. Oh, and you can't buy a petrol car because the car companies are under the hammer to sell EVs.

Just to be clear, the UK system is much simpler than the US system. There is just a bad law. That law could be repealed with a majority in parliament tomorrow, until it is repealed (spoiler it absolutely will not be repealed) the regulator can and will file these law suits. The best we can hope for is that the regulator (Home Office) just don't bother trying to enforce the law.

The core problem is the people writing the laws are know-nothing busy bodies who write crap laws and then cause massive problems, and we've demonstrated over the last 18 months that you can fire literally 70% of the UK Parliament, replace them all and still end up with the same rules written by the same know-nothing busy bodies.

I don't like this line of reasoning because it's largely just crystalizing a loss. We have this in the UK - houses of multiple occupancy. It's a great idea where you take a home that in the 1980s would house a family and split it into 5 flats where each person can rent 10-20 square metres each. I would much rather someone did something to address the fact that the average family in the UK can afford roughly 1/5th the amount of housing they could in the 1980s. And of course, because of this arbitrage now a family that wants to live in that home is competing with the rental income of 5+ tenants in a HMO.

Surely, the correct solution is just to put in some simple rules to bring the cost of housing down. For example: planning restrictions are suspended until the average family home hits 3x average family income. Rather than just packing us like sardines into ever more expensive houses.

Man I feel so dumb. Why haven't I been doing this in my job, if I could just see the commit that fixed my issue this would all be so easy.

Zuckerberg rushing into every new fad with billions of dollars has somehow tricked people into thinking that's what big tech is about and all of them should be shovelling money into this.

But actually every other company has been much more strategic, Microsoft is bullish because they partnered up with OpenAI and it pumps their share price to be bullish, Google is the natural home of a lot of this research.

But actually, Amazon, Apple etc aren't natural homes for this, they don't need to burn money to chase it.

So there we have it, the companies that have a good strategy for this are investing heavily, the others will pick up merges and key technological partners as the market matures, and presumably Zuck will go off and burn $XB on the next fad once AI has cooled down.

Yes and the result will be one of two options: option a (more likely) the underlying sites will literally just disappear, their business model no longer works and the content that you want (but apparently not enough to respect the authors) will cease to exist. It will most likely be replaced with AI slop replicas of the content you wanted. Or option b (much less likely) the content you want will move behind premium services where AI companies will have to negotiate subscriptions you will have entered the cable TV bundle era of the internet.

Perplexity has really convinced me about this. There is a clear difference between automated bots scraping data at bulk for later use, and automated bots working on behalf of users on direct requests. I can see a reasonable argument that some of the first type of automation could be tolerable for websites with strict limits, the second type I think by default should not be tolerated at all.

Perplexity's value proposition appears to be "we're going to take the stuff off your website, and present it to our users. We're not going to show them your ads, we're not going to offer them your premium services or referrals to other products, we're going to strip out the value from your content and take it for our users".

You can argue all you want about whether that's 5k impressions a day or 1m impressions a day. It should be 0 impressions a day. It is literally just free-riding.

Also, they're meant to be a professional company taking VC money to build a business, why are writing whiny posts like a teenager? The impression I get with a lot of these companies is that their business is losing money hand over fist, they have no idea how they're going to make it work and they look absolutely panicked as a result. They come across like a company I would want to be nowhere near.