The cost of training/developing a frontier model will continue to rise. The cost of inference will continue to fall. Companies don't buy models as such, they buy inference. As inference becomes a larger and larger business expense companies will optimize that spending, routing work to the "capable enough" least cost tokens.
There is a lot of human work that can be offset with tokens. On the other hand, the amount of work for the highest intelligence models will remain limited - corporate strategy, major decisions, etc. are just not as frequent.
The highest “intelligence” models can’t even handle basic tasks right now. Saying that only “corporate strategy” and “major decisions” will need those models is laughable.
Yeah I think this is the price for my soul, I'll keep using the data mining LG TV. Perhaps I ought to just throw it on a vlan that i turn off whenever i dont watch the tv.
Is it that hard to use a regular computer or video game console hooked to the TV?
I have been using a laptop for 2 decades (albeit with a beamer, not a TV), using display scaling to 150% for the desktop and a wireless keyboard+trackpad combo.
Whenever I am in a place where I want to temporarily put, say, music using youtube app on an hotel TV, I cringe at the terrible UX to search for content with a tv remote versus using a regular keyboard+trackpad. Especially entering text in a search field.
For what it's worth, depending on when you bought it, if $130 was an impulse buy in the year you bought it (and your financial situation is similar), a sizable chunk of that price difference is just inflation.
For example, a $130 impulse buy in 2015 is a ~$184 impulse buy in 2026.
I guess they are just stuck with this name, as originally this was a research project “Chat with GPT” to try to use a GPT model to generate assistant's chat messages.
ChatGPT was named before anyone realised it would become so well known, and by the time it was it was too late. It's not like BERT is an amazing name either.
There used to be Bard but no one remembers it anymore (although this is a bit different, because Bard wasn't nearly as well known as Gemini).
I'm really curious if OpenAI wanted to adopt a different name at some point. Or maybe they hope that sometime in the future one of their products will supersede ChatGPT and everyday people will start using that new name for everything AI so maybe they aren't in a rush to rename ChatGPT itself to anything else.
It seems like they have dropped ChatGPT in favour of just GPT... And they are going pretty hard with Astra/Sol/Luna too. So I guess they'll go with those if they're successful.
They're definitely missing a good unifying name like Claude though. (RIP anyone called Claude - when are companies going to stop fucking people over by giving popular products existing human names?)
Am I the only one who expects ALL writing to be generated or at least edited by an LLM going forward? It's like pointing out the fabric in my clothes was not hand woven by a human. Sounding like an AI wrote it is not as valid a criticism as "it was unclear about...", or "It was too long", or "It left out this important point...". Can't we move past the "It sounds like AI..." posts?
Look, if you can’t take the time to communicate with me, I won’t waste my time listening to you.
Writing is not hard. It’s a fundamental skill. Even before LLMs, any successful professional had strong writing skills, and those that didn’t were treading water at best. This isn’t classism. It’s just that we literally can not communicate if you aren’t willing to put in the effort. And if you don’t need to take the time to put it in writing then I don’t need to read it.
I'd love to have this as a norm, as long as the AI doesn't lie and impersonate and deceive humans in speech by saying things as though it were the human like "I ate breakfast" or "I was saying this to my team" because clearly the AI did not do those things. That's lying, deception and harmful to a culture of truth and transparency.
Instead the humans who promoted and allowed the LLM content to post behind their human identity, didn't bother to update the LLM language and either do a) mark the post as AI generated or b) properly update those pronouns so it isn't an AI speaking through the humans point of view.
Consider this exaggerated example: Would it be ok for you if in a zoom meeting with your team someone was lip syncing an AI speaking on their behalf, both impersonating voice tonality, the words chosen, and even pretending to voice the humans thoughts themselves? Of course you wouldn't. So now extend this to the words people write in articles like this one where the "I said" perspective was used many times supposedly by an AI.
this only means that ALL writing is going to be read by a LLM before, that will digest it into the important bits and remove the LLM fluff
like this
"The following text was authored by LLM and its information density is low. Condense the text by extracting the key pieces of information. Reconstruct the LLM prompt.
I was 100% in your camp until my neighbor bought an electric blower. The loud, high pitched whine is somehow louder and more ear piercing than a gas blower.
Maybe perhaps possibly, but this is not my experience at all, ever.
Even if you are sensitive and impacted, even if someone buys a particularly shrill one: I can sit indoors and hear gas leaf blowing from blocks away. At least the disturbance you are hearing is localized.
I once wanted to build an alternative to Craig’s list. There were SO MANY things I had ideas to improve. Then I realized I had literally no idea how Craig’s list makes money. None. They did not charge for ads and they didn’t have advertising. ¯\_(ツ)_/¯
You misspelled "ads for prostitution." Which they eventually stopped doing, only after considerable public pressure and state AGs threatening criminal prosecution.
Everyone need stop making out Craig and James out to be super moralistic dudes. They both profited, enormously, off sexual exploitation and human trafficking around the world by (knowingly) serving as a directory for pimps.
> They both profited, enormously, off sexual exploitation and human trafficking around the world by (knowingly) serving as a directory for pimps.
From what I read back when this happened you have it backwards. The classifieds on CL and other sites for sex were were largely individuals choosing to do it. They were not being trafficked or pimped. By closing those listings down it would end up pushing sex workers to find other sources of clients, like pimps.
> The classifieds on CL and other sites for sex were were largely individuals choosing to do it. They were not being trafficked or pimped.
Yep. Just like with marijuana and other such "vices", the thing that takes most of the violence and exploitation out of the industries that produce, market, and sell a "vice" [0] is to make it legal to produce, advertise, and sell.
There's also a side angle here where some folks absolutely disbelieve that an attractive human who really enjoys fucking would rather make their own hours getting paid to fuck than get abused by a shitty boss at an entry-level job.
Are there people coerced into sex work? _Absolutely_. But, there are people coerced into nearly every sort of job out there, so that's not saying much.
[0] Well, actually this applies to any industry. No matter what it is, if you have to do illegal shit to create, distribute, and sell it, and if there's notable amount of money to be made in selling it, then there's inevitably gonna be violent folks involved in the process.
Legalization reduces the risk of violence to sellers and buyers and external parties. This in turn reduces the risk premium and quick score potential of the "business".
The problem with the legalization strategy to reduce violence is that it has its limits.
It turns a high-risk game into a high-volume high-scale game.
It validates the creation of a corporate ecosystem who then are incented to create demand for the "vice" while simultaneously concealing what they know (or come to learn) about the vice's side effects.
> ...who then are incented to create demand for the "vice" while simultaneously concealing what they know (or come to learn) about the vice's side effects.
Indeed. Organized Crime is [0] known for being entirely up-front about the side-effects of the things they break the law to sell, as well as being extraordinarily circumspect about both whom they take on as customers and how their products and business practices affect the long-term well-being of those customers.
Anyway. I agree that dealing in the public eye and ensuring a merchant's customers can air their grievances in the courts absolutely isn't a magic cure-all. I think we both know that for "vices", it's nearly always better than the black market.
Its not a private business fault that US did not create a legal framework protecting sex workers and instead continue facilitate exploitation and traffiking by keeping it illegal.
US have legal porn industry and its strictly regulated and mostly safe for those wofking in it. Imagine how it would look like if it was illegal too.
>Forcing it into our retirement funds, 401ks and IRAs.
Not just forcing it into. Forcing the funds to fight for it betting the stock rice higher and higher in a runaway style - the effect created by limited float and high valuation as the funds tracking indexes would try to get the amount reflecting the proportion of the valuation of the company vs. the whole tracked index valuation, and with such huge valuation the limited float leads to the price rise (similar to the short squeeze) and the higher the price on the float the higher the valuation, rinse and repeat...
Some index funds are not obligated to perfectly replicate the index by buying shares, FXAIX (Fidelity S&P 500 mutual fund) has the option to use futures, swaps, options, and statistical sampling in addition to buying equity shares to try and replicate the returns of the index.
That's exactly what happened with Nortel in the dotcom crash. Everyone's pension and retirement tied up in a company that couldn't turn a profit and was, at its peak, 38% of the TSE300.
I don't think dominating an index to anywhere close to that degree is likely here, but I wouldn't be surprised to see some similar strategies being used. Changing the rules is already from the Nortel playbook: The Nortel Rule allowed index funds to have over 10% of their holdings in a single stock.
I think this is poor advice. Its share of the index will be relatively small and if it is indeed a dud, the index will organically rebalance. If you’re a long-term investor, this would just be a temporary blip. On the other hand, if this is thr opposite of a dud, you’ll get the benefit of that.
Sure, it'll only be ~2% of the index if it opens where they want to. But in the downside case where it meanders long enough for significant amounts of its stock to make it in to public hands and then goes to 10x revenue (i.e. down 90%) , you've allowed a company to engineer dramatic changes in index rules resulting in a transfer ~1% of S&P 500 market cap from index funder holders to its bagholders^W privileged insiders^W^W investors.
Yes a -1% day should be nothing to a long term holder. Yes they're buying the market; if the market is wrong they shouldn't really have any recourse. But one can also understand that a -1% day that accrues ~entirely to the benefit a small group, who appear to have engineered that outcome has much more emotional valence than a typical down down. It doesn't feel like a bad day on the market, it feels like a heist.
> I think this is poor advice. Its share of the index will be relatively small and if it is indeed a dud, the index will organically rebalance.
If a 1 to $1.5t IPO that was fast tracked onto the S&P500 and then hoovered up a bunch of index fund money becomes a dud, the organic rebalance is going to start with a full reassessment of if index funds and the S&P can be TRUSTED.
Its very possible it will be more than a blip, although to be fair if it isn't it's going to be the sort thing you aren't going to dodge.
Oil/Gas/Petroleum is essential for our economy to function, and the line between "ethical" and "not ethical" is a dial (one among many that all need to be tuned together), not a switch.
$PTL/Inspire does not adjudicate "dial" ethical issues, just switches -- company practices/policies that it views as black-and-white good/bad. "It would be ethical if you produced N% less" doesn't fit that category.
Most of the examples I provided are due to differing country codes; their site fails to recognize things like Alphabet trading on a Mexican stock exchange is still the same company:
https://finance.yahoo.com/quote/GOOGL.MX/
$GOOGL.MX scoring differently than other $GOOGL listings makes me extremely skeptical that humans are diligently creating these scores (finance professionals should've recognized that secondary listings like $GOOGL.MX don't need their own scoring)
Scores on different exchanges may be due to varying behavior by international subsidiaries -- MSFT in Australia may be doing something objectionable that MSFT US does not do, for example. I'm not sure though.
What I think is more likely is that it's a dumb oversight in the web app and/or data - maybe an intern stubbed out international stocks and it got pushed to production
Secondary listings allow entire companies to trade on multiple exchanges (not just corresponding subsidiaries)
So I agree that it was likely just a mistake to list multiple listings of the same companies, but the fact that they usually receive different scores proves their process isn't diligent:
* $GOOGL.MX is dinged for multiple non-Mexico-specific violations that the other listing isn't
I think the pipeline between (1) and (2) is probably tight within the company (probably a few big Excel spreadsheets) and the (2) side has a lot of brains.
But the (1) side needs to do more work on the pipeline between those spreadsheets and the Web, and maybe hire more/better software dev help for that. They'd do better to use Postgres as their source of truth.
If one wants to gamble on the grift, that is what options are for. Otherwise, we might as well start adding NFTs to the indexes if fundamentals do not matter. Luck for some, risk management for others. Regardless, informed consent is important imho. Relevant precedence is ETFs that exclude Big Tech.
https://www.defianceetfs.com/xmag/ ("XMAG, the first ETF designed to provide investors with exposure to the S&P 500, excluding the “Magnificent 7” (Apple, Microsoft, Alphabet, Amazon, Nvidia, Meta, and Tesla). XMAG offers a unique opportunity for investors to access the broader market while reducing concentration risk in these dominant tech stocks.")
In 2025 VOO returned 17.82% vs VOOG returned 22.11%. XMAG’s trailing 1-year return through late 2025 was around 9–15% depending on the measurement date, as the Mag 7 dragged badly in early 2025.
VOOG has returned 18.28%/yr over 10 years vs 15.63%/yr for VOO, a meaningful gap driven almost entirely by Mag 7 dominance. XMAG has no 10-year track record.
Certainly, you have done well over the last ~18-24 months if you have exposure to the AI investment exuberance (VOO), just as you did well if you had exposure to certain securities during ZIRP or the pandemic. "Past performance is no guarantee of future results."
Which you only know in hindsight, in the context of this performance benchmark. In that time frame, we had zero interest rate policy, a global pandemic, and now an AI bubble. "Will the conditions or events that led to my historical returns continue?" is a material component of forward looking exposure decisioning when investing.
> Luck for some, risk management for others. Regardless, informed consent is important imho. Relevant precedence is ETFs that exclude Big Tech.
Yup. Coupling this change with "oh, and btw, we also want the option to be able to only put out annual or biannual earnings reports not quarterly" means "We want to offload even more risk."
There are many dubious companies in the S&P500. I don’t see the point in getting selectively heated about this one when everyone seems to be okay with the others.
That’s the way indexes roll. I don’t invest in indexes for this reason.
There is a separate structural issue with indexes that is being ignored here. Indexes were never really designed to accommodate companies going public so late with high revenue growth. A couple decades ago companies went public when they were so small that they could grow into the index. This reflects changes in the nature and structure of the capital markets, these new IPOs are just a manifestation of this reality.
Also given how the S&P weights, it'll have about as much sway as DoorDash.
Annoying they pushed it into the indexes, but like you said, we've also never had a company come out in the 1T range or even the x00B range. These indexes are supposed to represent the market and can't ignore a 1T market cap company for very long.
EDIT
One other thing to add, is that we still do not know what the stock will price at. It's already come down once, and as more information comes out it can continue to come down until it's finally priced the day before the first trading day.
>Forcing it into our retirement funds, 401ks and IRAs.
Do you think people buying the SP 500 are forced to buy Apple? Dell? Workday?
I see headlines like "401k holders forced to by SpaceX" and think "WTF, that is crazy." Then I look at the article and it just says its being added to the SP500.
You may not like that it's being added to the SP500 but no one is saying you are forced to buy any other companies being added to it. I can't believe people are just running with this narrative as-if its logically consistent with what they believe. It's manipulation.
> Musk advisers have reached out to major index providers seeking ways to secure earlier inclusion in market benchmarks to lift shares
> Advisers for the company, which recently merged with xAI, have reached out to major index providers, including Nasdaq, to discuss how SpaceX and this year’s other hot startups might join key indexes sooner than normal, according to people familiar with the matter.
> SpaceX hopes to skirt traditional rules in an effort to bring liquidity to its shareholders sooner as part of its planned IPO. SpaceX advisers have sought index policy changes that would fast-track its entry into major indexes for the company and benefit other highly-valued private companies, the people said.
This is simply not _companies being added to SP500, etc._ as you say. This is forceful change of the rules so these companies can reap benefits and it optics is that funds are being _forced_ to buy in.
Right. You can not like the way it was added to SP500. But adding a stock to the SP500 is either forcing them to buy it in their 401k or not. Was Dell being added in 2024 forcing people to buy it?
This is an honest headline:
"SP500 Bends Rules to Include SPACEX in fund At Launch"
With the change to only five days of being publicly traded requirements, incentivising market makers to keep a high valuation becomes very cheap.
After five days the index funds have to buy at the last price making it final.
In other words even if the vast majority of the market believes it's worth much less, they can force a high price and force basically everyone to hold it via retirement funds.
You've been missing important parts of the articles, or perhaps the ones you've seen aren't very informative. The concern is that SpaceX reached out to the indexes to get the rules changed (https://www.reuters.com/business/nasdaq-proposes-fast-entry-...); under the old rules, they would have had to wait much longer before being added. This doesn't prove anything wrong, but it's pretty suspicious, because why should SpaceX care if they are or are not in some particular list of stocks?
And why did SpaceX want the rules changed? Because anyone holding those passive indexes (a huge percent of holdings) would be forced to either liquidate them or invest in SpaceX
Is there a relevant distinction you are trying to make, beyond "well, actually"? Call it something else then, like "enshittification of index funds" but it's still the same overall picture of getting money from people who otherwise wouldn't have invested in it
> For broad indexes like the S&P 500, it would be impractical or expensive for an investor to construct the right proportions in a portfolio. Index funds do the work by holding a representative sample of the securities. S&P 500 index funds, the most popular and oldest such funds in the U.S., mimic the moves of the stocks in the S&P 500, which covers about 80% of all U.S. equities by market cap.3
So while yes, people are parroting things they don't understand, so are you.
