He spent that much time and you still misunderstood the direct message and missed the subtext.
The lie is coding is solved, the proof is they had an outstanding coding issue they were working on for over a year while saying coding is solved. There’s a great number of other issues with their own software that disprove their premise, but you only need one counter example to disprove something.
And because you missed it, the subtext was they want you to use loops not because they work but because they burn lots of tokens thus making them more money.
Not quite, current nighttime load is largely a function of cheaper nighttime rates. People don’t set their EV’s to charge from 11-5AM because that’s the only time their cars are plugged in. If rates crater at noon on Sunday, there’s many an EV happy to suck up power then.
So yes batteries are going to continue to grow rapidly, but it’s a smaller role than it might seem.
Great, so now not only is power production nondeterministic, your cars tank also is. It was too expensive and the algorithm decided to wait charging, so no spontaneous road trip for you, sorry.
As an electric car owner, this is absolutely the biggest non-problem ever. If you're planning a long journey, you push one button at some point over the preceeding week and it charges to full regardless of price.
There’s nothing non deterministic about cheaper daytime rates as you scale solar production. Net result, lower average electricity prices but a slow rise in nighttime rates across decades.
Similarly people respond to price changes, that’s the foundation for how capitalism functions. You don’t need to care, but many people will choose to save money when possible.
I mean, assuming you don't zero your charge out when returning home, you could just take a few minutes to use a rapid charger part way through the journey...
Total electricity produced by coal + gas is down over the last 20 years. Total electricity production is up, the difference is from wind and solar.
This administration swapped to actively suppressing Wind and Solar via tariffs etc, and yet the trends continued because the underlying economic reality heavily favors battery backed solar.
I think that's part of what's notable about this. The administration hasn't been able to reverse the trend despite putting a massive thumb on the scale against projects like offshore wind and tariffs on solar panel imports.
There's probably a delay in the effects though since projects started before they took office are probably starting to thin out and finish up. We'd have to look into the permitting of new projects or wait for to see how big the decline in new capacity turns out to be in a couple years.
A lot of comes from state initiatives too. Texas being conservative also happens to be very pro solar. I’m in the business and we have some great projects there. The US military is also pushing solar at their facilities. Then you have many private-state partnerships like tolls investing a lot in solar. The outlook in general is pretty positive in the US, a lot more than what people would think.
Anyone who still even views this as a conservative/liberal issue, is someone who is in the pocket of the fossil fuel lobby. Solar is simply a very cheap and realiable way to generate electricity. Much cheaper than gas and coal nowadays. Pure economic incentive is going to continue to drive its adoption.
As someone from another continent, I giggle when I see doomsday-prepping anti-govmint fiercely-independent cowboys hating EVs and loving gas.
Gas has a 6-month shelf-life, and is attached to a whole geopolitically volatile military-industrial complex. Meanwhile an EV + solar can be actually self-sufficient and last for a decade or two. A realistic Mad Max would have been EV battles over solar panels.
this’ the other Max from China, where they be scavenging leftover panels. The Mad Max we know surely happens somewhere else, and perhaps near Texas given all the participating parties…
I also recall a New York times article from many moons ago suggesting that a lot of Texas oil wealth got repurposed into a large-scale wind energy infrastructure, but my info might be out of date.
It's sort a "broken clock right twice a day" thing, but I agree with not doing offshore wind in the US.
The divergence immediately follows in that I wish they would just push onshore wind.
It's sort of a circular issue, it's madly expensive because we haven't built a lot and aren't super good at it, and we don't get much of it built because we aren't great at it and it always is ludicrously expensive.
The US has a uniquely underdeveloped maritime sector, we don't build a lot of the massive turbines you use offshore.
You drive through central and west texas, it feels like there might be more wind turbines than people.
We've kind of already made the decision based on what works.
Ocean winds are strong and predictable in ways that are really beneficial to the wind farms so the extra costs are balanced out by the fact that there's always a strong usable wind to harness too.
Also the "we're bad at this because we don't do the so we can't do this" is throwing away a great project and solution because of a temporary problem. Once we start doing it in significant numbers we'll rapidly get better at doing it too.
Like for many kind of technology the are both advantages and disadvantages to offshore wind farms.
"Advantages:
Offshore wind speeds tend to be faster than on land.1 Small increases in wind speed yield large increases in energy production: a turbine in a 15-mph wind can generate twice as much energy as a turbine in a 12-mph wind. Faster wind speeds offshore mean much more energy can be generated.
Offshore wind speeds tend to be steadier than on land.1 A steadier supply of wind means a more reliable source of energy.
Many coastal areas have very high energy needs. Half of the United States’ population lives in coastal areas,1 with concentrations in major coastal cities. Building offshore wind farms in these areas can help to meet those energy needs from nearby sources.
Offshore wind farms have many of the same advantages as land-based wind farms – they provide renewable energy; they do not consume water; they provide a domestic energy source; they create jobs; and they do not emit environmental pollutants or greenhouse gases.2
Disadvantages:
Offshore wind farms can be expensive and difficult to build and maintain. In particular:
It is very hard to build robust and secure wind farms in water deeper than around 200 feet (~60 m), or over half a football field’s length. Although coastal waters off the east coast of the U.S. are relatively shallow, almost all of the potential wind energy resources off the west coast are in waters exceeding this depth.3 Floating wind turbines are beginning to overcome this challenge.
Wave action, and even very high winds, particularly during heavy storms or hurricanes, can damage wind turbines.1
The production and installation of power cables under the
seafloor to transmit electricity back to land can be very expensive.1
Effects of offshore wind farms on marine animals and birds are not fully understood.4
Offshore wind farms built within view of the coastline (up to 26 miles offshore, depending on viewing conditions5) may be unpopular among local residents, and may affect tourism and property values.3
I think the idea of offshore wind is nice. There is a lot of ocean out there and by putting the "ugly" (I don't mind them.) turbines out of sight we get the best of both worlds.
But the realities of the idea - the engineering - is problematic.
The ocean is a harsh environment and maintaining something deliberately put out of the way in a harsher environment is far more expensive.
There are a lot of ocean, but limited shores. I guess that's the only economical place to build it, but of course it blocks scenery. If people truly built it in open ocean, less people will complain.
That’s ~true for now but keep the date someone said it in mind. It’s really easy to keep thinking technical limitations still true when they are wildly out of date.
Tariffs on solar panel imports should stimulate domestic solar panel production, but only when they are high enough and applied long enough to justify investments into new solar panel manufacturing facilities.
"The research results show that China controls the supply of primary materials, manufacturing, installed capacity, and recycling capacity. China alone produces at least 80 % of the main components of PVs."
I'm asking about the supply chain for drilling, refining, and processing oil, my friend. Is that supply chain completely domestic? Nothing from China in there?
You would need to do full supply chain analysis. Probably only one doing full supply chain analysis is Pentagon. They had experts for this kind of work.
"How The CIA Secretly Bought Soviet Titanium To Build The SR-71 Blackbird"
You can do it with targeted tariffs with assurances they'll last but these broad tariffs make it harder to get the base materials you need to build the panels out of in the first place plus they're so crazy they're almost guaranteed to be wiped out in a few years if not sooner so as a manufacturer they don't have the confidence that the cost equalization of the tariffs will be around long enough to not bankrupt them.
Another way to do it would be guaranteed buys for electrifying military etc and grants for projects using US made cells instead of foreign ones that could also effectively subsidize local production.
It's like a lot of things done by this administration they do it such hamfisted and obvious ways that they don't accomplish their nominal goals. See a lot of the court cases where they've been blocked in implementation because they said the quiet part out loud. eg: it's usually REALLY hard to prove malicious prosecution but they keep saying out loud "we're prosecuting this person in retaliation for their protected activities".
It's important to tariff the basic materials Polysilicon, wafers, if you want to spread solar production around the world, not only in a single country.
It's especially notable because there isn't just the thumb against offshore wind, solar panel tariffs, and even EVs. Chinese EVs can't be imported because of tariffs and many conservative states a pretending that EV drivers "don't pay their fair share because they don't buy gas" - except most gas taxes haven't been adjusted in multiple decades and don't even begin to pay for the cost of maintaining roads. Fuel taxes are a tiny portion of any state revenue.
There has always been a massive thumb on the scale in the form of tax breaks, direct subsidies (billions a year alone on this), land leasing, etc for fossil fuels and their use. Favorable public policy. And what the IMF calls implicit subsidies - the cost of impact on the climate/environment and people's health.
When a refinery is pumping out pollution and everyone in the area is getting sicker than people in similar areas - that costs us as insurance ratepayers and taxpayers.
Unfortunately this is one of the few cases where both sides are close to the same -- they both chose to heavily tariff foreign EVs. Something to remember when Democrats talk about climate change.
"closer to the same" except one side has consistently promoted the building and sale of EVs in America.
If you were feeling generous you could credit then with the entire existence of modern EVs given their support of the nascent industry for decades in California.
But sure, let's focus on the mote in Democrat eyes and ignore the insanity across the aisle. That's what got us into this situation, so why stop now.
It's the same in the UK. The CfD "subsidy" mechanism for solar power results in the solar farms paying us money almost every day - including today - but you will still see right wing politicians vow to eliminate this "subsidy".
I doubted what you wrote, but everything you said is correct (for the last 10 years, at least). Over the time period, natural gas increased 740 TWh/year (to 1870) and coal decreased 940 TWh/year (to 650). Electricity production is up ~7%, but that's quite low compared to the growth of everything else.
> This administration swapped to actively suppressing Wind and Solar via tariffs etc
Biden's administration put on solar tariffs, but of course I'll grant the current administration is fucking up everything else possible.
Plenty of blame to go around, my understanding of the timeline is:
Trumps first administration put in solar Tariffs with China (25%), Biden administration increased them with China (50%), 2nd Trump administration increased those and applied solar Tariffs to other countries. Though honestly I’ve largely stopped paying attention at this point.
I think the issue is that the tarriffs just don’t really matter anymore because the panels are so cheap that they’re dwarfed by the “balance of system” costs — installation, racking, inverters, cabling, etc.
It matters because of opening up cheaper ways to install them. For example, in some Scandinavian country they're cheaper than fence panels, so they just use them for fences, and the power they generate is a nice bonus.
As a foreigner it just seems.so braindead that the administrations would tarrif solar panels. The US doesnt have a great manufacturing capacity for solar panels compared to established manufacturers. The high cost of new production ensures slow uptake.
From a place that embraced solar rebates, and has subsequently benefited from having in place solar battery rebates, we have a thriving industry of solar installers, electricians, and an ever increasing amount of local grid energy security in the event that storms or accidents cause supply disruptions. About 5% of households will likely not see an energy bill for the next 20 years. Another 40% have solar that covers daytime energy requirements.
The requirement for baseline coal.and gas has been decreasing - though will not completely abate.
I live in a state that produces abundant coal for power and steel. We have decreased our carbon emissions to 35% below 2005 levels.
Either you want to secure the full energy chain or you do not. There are many comments in this thread about how the solar panels come from China. If that bothers you, there needs to be some financial reason for a domestic factory to open.
> Either you want to secure the full energy chain or you do not.
Is every part of the fossil fuel production chain "secure"? All oil production and processing equipment, chemicals and so forth are produced domestically?
By that token normal reproduction qualifies as immortality as parents share a lot more DNA with their kids than those cell lines with their originator. Biologically those cell lines are generations of new organisms not the same individual living indefinitely.
Further those cells can’t compete out in a forest somewhere. Going multicellular means optimizing in ways counterproductive to survival on their own. But failing to cull off defective cells means you can’t have a functioning multicellular life form, thus making this a dead end.
So no in nature there isn’t some switch that means those cells get to survive indefinitely. The closest viable option is as an infection shared among very closely related organisms, but that’s not a stable option which means it’s really rare in nature.
Executive compensation is often based around share prices, so this can be worth quite a lot to the people making these decisions without any long term upside for the company.
If you want to understand how companies behave you really need to look at things from the perspective of people making the decisions.
I don't see Alphabet share price changing much just because of SpaceX being valued 2T instead of lets say 1T (being extremely generous). In fact this deal will hurt their profits, which is more likely to hurt Alphabet stock price than the valuation of an asset that they hold.
> I don't see Alphabet share price changing much just because of SpaceX being valued 2T instead of lets say 1T
Half of Alphabet's revenue increase last quarter came from marking up unrealized gains in their Anthropic investments.
I'm not saying Alphabet is doing this to juice the share price, but I want to point out that they don't have to sell shares to post banner earnings results and see a 10% jump in share price overnight.
That’s the brand not a specific device these days Apple is selling well over 200 million phones a year.
The original iPhone sold 6.1 million units in its lifetime and topped out at the 3rd most popular phone in the US, being dramatically less popular in Europe etc. IE less than 2% of current annual sales.
By comparison just the base model iPhone 15 was the most popular phone globally for 3 quarters excluding the iPhone 15 Pro Max and iPhone 15 Pro which where close. Part of that is just being a way better phone in terms of battery life and being significantly more resistant to water etc.
That’s a simplified and somewhat outdated version, there’s a huge range of technology to mitigate each of these issues but infrastructure generally isn’t free. A turbine does provides inertia for free where you need to pay for a flywheel or battery system.
In the end a renewable heavy grid can be extremely resilient due to all the batteries, but smart battery systems need to be incentivize or mandated because ‘dumb’ batteries are cheaper.
That assumes Tokens will remain a meaningful expense. I’m not sure developers will find uses for ever more tokens nearly as quickly as the prices fall.
How are we so confident that prices will fall? Isn't the exact opposite happening, right now, during arguably the most critical part of this whole saga (pre-IPO to make things appear as beautiful and as not-obviously-illegal as possible)? And the only reason they were "falling" previously was for hyper growth.
The Growth aspect mentioned is that VCs are subsidizing the bill right now, so it is hard to know if at the current moment the demand curve would promote as much usage without it, but assuming demand remained constant (not even growing), you could expect token prices to be competed down. It is a commodity without a moat.
Now that we have pretty decent open source models, anyone can create a new business to supply more tokens. Sure there’s short term scarcity: energy, GPUs, cooling, but this is a scale up problem. More token demand = more data center build = more energy plant build. This downward pressure will also keep frontier private model prices in check.
Differentiation seems to be happening at the harness level, whereby we can expect token spend to be a metric to compete on and drive down for the customer (at least hoping tools in the application space don’t continue token based billing as their primary revenue stream).
These are not short term hyper growth forces, but a fundamental alignment of incentives.
Pricing on SToA models probably won’t fall, there’s no reason for the frontier labs to lower their prices.
But we’re seeing lots of open weight models that are either pretty close to SToA, or more importantly, perfectly capable of doing all the low level token insensitive grunt work when writing code. Pairing them with SToA models for long horizon task management, and you’ve got a very cost effective system.
The frontier labs have put little effort into cost efficient inference, they don’t need to, but folks like DeepSeek clearly are, and have achieved some impressive cost improvements. Given DeepSeeks models give you 70% of the capabilities for 30% of the cost, expect people to start moving lots of workloads to providers that provide cheap inference for open models, and huge competition to appear to provide that cheap inference. It’s truly commodity LLM inference.
In turn expect more companies to focus on building inferences efficient models, because someone that can build a model that provides 70% of SToA capabilities for 10% of the token cost, immediately eats up huge amounts of the available inference market.
Another factor in all this, is it’s becoming increasingly clear that building custom agents/workflows for LLM to operate in, is required to get the best out of these models. That means people are implicitly building the infra needed to use multiple model types and evaluate workflow performance end-to-end. Which in turn means they have everything they need to plugin in future, cheaper, inference providers and quickly evaluate if they can change their model provider.
it is falling if you look elsewhere, deepseek made their 75% discount on their V4 models permanent, on one hand there's LLM improvements that make inference cheaper (e.i. MoE, hybrid attention), on the other hand we're getting more inference focused chips that break the nvidia monopoly.
i don't think a lot of people know this, but a cluster of GPUs can serve multiple clients without much of a drop in performance, e.i. worst case scenario you band together with 6-16 people to run a 2-3 H100 server to host deepseek V4 Flash or 4-6 to run Pro, and you're getting the same performance as if you ran it alone, this means a lot of companies can afford throwing 50-100k into their own LLM server cluster.
We're at a price point where if you push it further people will move, there's no real vendor lock in, your agent config, skills, MCP servers etc are all reusable with other models and harnesses, so unless you get all providers to collude on a price hike, you risk an exodus of customers
In the one direction the hardware continues to improve, new buildouts continue to come online, and methods for improving the parameter efficiency of models continue to be discovered.
In the other direction models continue to grow larger, new customers continue to arrive, and existing customers continue to find ever more creative ways to burn large quantities of tokens as the prices fall.
I doubt anyone can say with certainty where the equilibrium will be 1 or 5 years from now largely because (among many other things) it's impossible to predict how much of the current economy AI will end up eating. In general though the third party providers of open weights models are probably the most reliable data source available since they have little to no incentive to subsidize usage.
Prices have fallen dramatically over the last few years. It’s just that our standards have increased because we are using AI in ways that were not possible with worse models. But for the same level of “intelligence” as we had a couple years ago, the prices are so much lower.
The bloat was a result of withdrawing government funding / oversight and replacing them with student loans.
Looking like the DMV was no longer accepted as you needed to actually attract students, which means supporting an ever wider array of different things. Daycare, job placement, etc etc individually get tacked on and before you know it administrative overhead has gone wild.
You see similar effects with housing where people buying stuff with major loans are simply less cost conscious. Saving even 1,000$ on a kitchen is a big deal for most people if you need to write a check for it today, split it over 30 years and suddenly it seems largely irrelevant.
The lie is coding is solved, the proof is they had an outstanding coding issue they were working on for over a year while saying coding is solved. There’s a great number of other issues with their own software that disprove their premise, but you only need one counter example to disprove something.
And because you missed it, the subtext was they want you to use loops not because they work but because they burn lots of tokens thus making them more money.
reply