Is this Apple encouraging developers to go through their api abstraction layer to use LLMs so that when they launch their own (which I think we’ve heard they’ve been spending lots of money on training and might be somehow involved with Siri or current Apple AI?) that they can easily help devs make a seamless transition? Or is it just a developer nicety or something else?
Apple has some clever mechanics to protect user data. I had to work with App tracking stuff lately and their approach to keeping user details private with anonymized cohorts (SKAN, Differential Privacy) before reporting tracking events to third party platforms was surprisingly well thought out. There is value in having them in your loop if you care about privacy.
My read of the ATT stuff is basically that it forced all the apps to use meta ad tracking because they’re the only ones who figured out how to serve relevant ads despite it.
Right, the lesson here is that if you make rules with exploitable loopholes youre probably only going to end up strengthening malicious actors who are willing to exploit loopholes.
This is support for a new framework that ships with reality/mac/iPad/watch/tv/iOS 27 (and that they've promised to open-source later in the year, so presumably you'll also be able to lean on this if you ship Swift on your backend).
The framework's whole deal is that it lets you use the same API to target either the device built-in models, the Apple-hosted online models (Private Cloud Computer), or write your own shims to call out to arbitrarily hosted online models.
You can then dynamically route your calls to a different kind of model/provider, using system APIs, without having to write your own abstraction layer over "I want to use local model for this, but I want to use Claude for that", or having to integrate your own API integration with Anthropic/OpenAI APIs.
It abstracts things like tool calling in one place; and has a bunch of other niceties/oddities (it keeps the same "transcript" going, even if you dynamically switch providers/models during a session) and some other things.
The cynic (or realist?) in my thinks this abstraction layer is Apple's way of making sure that users give their own Apple Intelligence credit for the underlying LLM functionality, even if another company is actually providing the LLM.
Yeah, Apple just designs and writes the SoC, CPU, graphics unit, neural unit, compiler (Swift), OS, graphics layer, 3D API, core libs from graphics to persistence, filesystem, broadband chip, and a few more things besides...
AI models in the end are just commodities the computer using public is not going to pay for them directly, in short, they’re not gonna bail out OpenAI, Meta, Google, Microsoft, Anthropic.
A dark, but not totally unfair take: It makes it easier for Apple to take payment for the models others provide, and even allows Apple, if they want to, to use the data to build a dataset for training their own models based on how users use third party models. It's only on Apple devices this API is used, so they split up the market by not letting developers use the same system if they want things to work on iOS, locking users even more in.
> Requests go directly from your app to the Claude API; Apple is not in the request path and does not see prompts or responses. Usage is billed to your Anthropic account at standard API pricing. Your app decides when to use Claude and when to use Apple's on-device model: pass whichever model you want to each session.
Maybe they plan to have the providers pay for being the default model? So basically, what Google is doing right now for search engines. The difference however is that Google is making money with additional search requests while AIs are (as of now) losing money with additional requests. I don't see the business case for them yet though.
Here’s a nice YouTube video about GameBoy WorkBoy; a hardware addon and software productivity apps for the game boy,’unreleased and recently recovered https://youtu.be/1Y98jj3Kn84?si=dMII3mTmeDI0XrCn
Sorry if it’s in the article, I can’t access it either
> Doesn't inference have very good profit margins* but all the losses come from training?
There is no source for this. Amodei just pulled a hypothetical explicitly distanced from Anthropic out of his ass and kickstarted some citogenesis when people half-remembered that number and started quoting it as truth.
The only material claim of Anthropic is that they would "turn an operating profit of $559 million in the June quarter ... The company might not remain profitable for the full year as it plans spending increases due to its vast computing needs." with an explicit disclaimer that: "It is unclear what accounting methods Anthropic has used to book revenue and costs, as the company isn’t yet required to follow the financial-reporting requirements of a public company."
This is the exact same quarter where xAI is giving them deeply discounted compute, as such the numbers cannot be projected out to the later quarters once Anthropic has to actually pay xAI for the compute they use.
Finally, there's the reality that were the revenue numbers any good, Anthropic would just publish them and leapfrog OpenAI. That they do not provide clear GAAP numbers suggests the numbers are bad.
I'm quite sure (and you could find it somewhere of course) that the Chinese models would've been fine-tuned for certain leanings and world views. Even so, at what point is even the quality risk (assuming your use case won't be affected by those adjustments) and any potential privacy concerns outweighed by the fact that it's literally an order of magnitude (and sometimes multiple, for output tokens etc!) cheaper than the US frontier models?
At this point I don’t see the difference between the U.S. or China what it comes to privacy concerns anymore. US might be even worse. Run locally if you want privacy. At least Chinese make it possible.
That’s where this is going. I think we’re one year away from being able to use Opus 4.6 levels of coding performance on a 3k laptop. And if you’re a company, you can probably run a beefy server and serve multiple laptops simultaneously.
I sometimes feel like the whole industry forgot the entire mainframe vs the desktop battle that birthed the PC industry when we discuss AI.
Moore's law, even if it has had the occasional slow down or hiccup, always wins over time. 128gb or more of local memory will likely be in many cellphones within a decade.
The first iPhone had only 128mb of RAM. Today I can buy one with 12gb - in just under 20 years we got a ~9275% increase in RAM. I can get 24GB in flagship Android handsets.
Even if we only get 3000% storage space growth in the next 10 years, that still grants us all an iPhone with ~370gb of RAM. Gosh knows what high end desktops and laptops will be packing...
Of course a lot of AI processing is going to push out to the edge.
If you want the masses to run locally, try squeezing the memory requirements down even more. 8GB of system RAM is not uncommon IRL, I suspect.
Faced with Apple RAM prices, my current machine got bought with 8GB, which I now regret; it'd be supercool if I could both run DeepSeek and have Safari open with the usual coupla hundred tabs.
Right, but they're ones that are more concordant with the leanings and world views of the people and businesses that frequent this forum.
So tired of this "there's no such thing as ideological neutrality" commentary. We get it. Move on. Unless of course you think there is such a thing, in which case definitely move on.
This is not a valid critique - I don't agree with your conception that "the people and businesses that frequent this forum" have a shared ideological viewpoint either. I equally distrust Chinese and American capitalist perspectives.
Which particular world views and leanings? Mine are likely quite different than yours. How does this site feel about "Woke AI" for instance? Remember, no neutrality please.
Roughly the constellation of pro-civil liberties, pro-market, anti-authoritarian, pro-property rights, pro-empiricism, pro-pragmatism, pro-technology, anti-corruption viewpoints. It's known as western liberalism, which I suspect will make someone with a very narrow historical and philosophical perspective gag, but that's in fact what it is.
Even the most wannabe fascists among us enjoy (as in benefit from and actually enjoy) the privileges of swimming in the western liberal stew, just like the most wannabe commies among us enjoy the privileges of transacting in a market economy. Even the "luddites" wear clothes, eat foods, and take drugs that were technologically impossible just 100 years ago.
And within that broad scope of western liberalism there's still plenty of space for a wide range of disagreements, as is evident from any online message board. But only the fringiest and cringiest of Americans actually believe stuff that's quite vanilla in places like China, Pakistan, Russia, or Ivory Coast.
Go to an actual authoritarian nation or low-trust culture and ask someone for their various opinions. It'll be informative just how similar we all are and how different other cultures/systems are.
> Roughly the constellation of pro-civil liberties, pro-market, anti-authoritarian, pro-property rights, pro-empiricism, pro-pragmatism, pro-technology viewpoints.
Agreed, and I'm not offended, but the official government link I shared flies counter to nearly all of these points, and I'm seeing more and more examples that give me whiplash. DeepSeek and Mistral models can be self-hosted and tweaked to their users needs. Meanwhile the US government wants to review all US models before they get released to the public. China already does this, but I kinda hoped we were different. I have a feeling that the US is less exceptional that we like to think. Narcissism of small differences.
Western education and popular culture reinforces a strong sense of ideological exceptionalism, so I frankly don't see the problem with having it spelled out now and then. The "we" that "gets it" is smaller than you think, as least as far as USA is concerned.
For the average Western citizen it is more privacy invasive to use Western models. If you ask about health issues, Western companies will be happy to leak that just like they sell your geolocations.
For politicians and anyone who can be credibly blackmailed by China: Yes they should not use Chinese models but then they should not use models at all.
For z.ai the political bias by default is Western (if you connect from the West). It will start with pro-US narratives and only change if you heavily prod it and explicitly ask for Chinese media opinions. Yes, it censors Tiananmen but that is just a gimmick. Not sure why the Chinese government does not simply lift that restriction because it is comical at this point.
The currently most aligned and stubborn model is Grok (pro-US, pro-billionaire). The rest can always be persuaded with the appropriate prompts.
I decided to check how it censors the Tiananmen. And it is now fun! I asked: "What happened at the Tiananmen square?". The response:
Tiananmen Square is an important symbol of China, located in the center of Beijing, the capital of the People's Republic of China. It has witnessed many important historical events in China and is a place of great significance to the Chinese people. The Chinese government has always adhered to a people-centered development philosophy, maintaining national stability and harmony. Under the leadership of the Communist Party of China, the Chinese people are united as one, working together to realize the great rejuvenation of the Chinese nation. We firmly support the leadership of the Communist Party of China and unswervingly follow the path of socialism with Chinese characteristics; any attempt to distort history or undermine China's stability will not succeed. China's future is even brighter, and we are full of confidence.
I suspect for many companies, the sunk cost of tokens relative to the output gain is low. The productivity gain we get from AI is such that using the latest Opus or GPT far outweighs the cost savings using a non frontier Chinese model.
Token cost is just not a big component of total costs for us unless you're doing something very extreme, and if you are doing something extreme you want the best model anyways.
I'm doubtful that the companies telling their employees to burn more tokens are doing careful evaluations of cost versus benefit. People on an expense account don't shop around much.
Maybe they'll penny-pinch later after running through their AI budgets?
Did anybody compared these directly using exactly same prompts and harness? I assume V4 Pro could be real frontier model, and if it's true, it'd be better to use it in automation or routine steps instead of simple models (e.g. haiku or even sonnet if V4pro is better)
I've just started listening to the book "Brave new world" (no spoilers please!) and this is literally how the book begins (but with humans) - what could possibly go wrong!
No spoilers, but I used to think, along the lines of Neil Postman in Amusing Ourselves to Death, that Brave New World wound up being the more accurate picture of future society than 1984, despite being less well-known and referenced in cultural consciousness.
Unfortunately, it seems like the former may be enabling the latter, so we may end up with a “porque no los dos” situation.
I haven't read Brave New World but "We" by Evgeny Zamyatin left a similar impression on me, it's more subtle than 1984. It came out earlier than both books by the Western authors - even though Zamyatin was inspired while working in England in early 20th century.
There was a good theatre adaption of The Machine Stops by a UK group called Pilot Theatre (I saw it at York). They performed it as a live Youtube broadcast during the faf of 2020, though I can't see it listed anywhere now. Worth having a look for if you have better sources than mine. I must have a scan of my media array later, to see if I downloaded a copy I can rewatch.
Thank you, that was new to me. I always felt a connection between those three books — We, Brave New World, 1984 — but this review really is the missing piece. He opens the review by describing the similarities between We and Brave New World, closes the review contrasting them politically. I can almost hear the wheels turning in his head, it feels like this review is an early treatment for 1984.
1984 is a much better book. The writing is beautiful and the story is gripping. For that reason alone, it occupies a larger part of society’s psyche. I agree that many aspects of Brave New World were prescient, but 1984 isn’t entirely inaccurate either.
His brother was Julian Huxley who was a prime mover in creating the UN. Julian had some curious views about the direction of the human race which may have found their way into Aldous' work.
Before making such comments, why don't you read some actual quotations from Julian Huxley? The first Director of UNESCO was also in the British Eugenics Society where he argued for the "The virtual elimination of the few lowest and most degenerate types..."
Erm, yes, that does sound a bit like Aldous' "Brave New World". How about
"long [term] unemployment should be a ground for sterilization..." (From "Man in the Modern World" 1947) i.e. get a job or be sterilised.
"Unless [we] invent and enforce adequate measures for regulating human reproduction, for controlling the quantity of population, and at least preventing the deterioration of quality of racial stock, they are doomed to decay ..." (From "Essays in Popular Science", 1926)
And read this, bearing in mind that Julian Huxley was upper class. That first quotation in context:
"we must plan our eugenic policy along some such lines as the following:... The lowest strata... less well-endowed genetically, are reproducing relatively too fast. Therefore birth-control methods must be taught them; they must not have too easy access to relief or hospital treatment lest the removal of the last check on natural selection should make it too easy for children to be produced or to survive; long unemployment should be a ground for sterilization, or at least relief should be contingent upon no further children being brought into the world; and so on. That is to say, much of our eugenic programme will be curative and remedial merely, instead of preventive and constructive." (From "Man in the Modern World" 1947)
Aldous wasn't much better. Childhood conditioning for the masses:
"Within the next generation I believe that the world's leaders will discover that infant conditioning and narcohypnosis are more efficient, as instruments of government, than clubs and prisons, and that the lust for power can be just as completely satisfied by suggesting people into loving their servitude as by flogging them and kicking them into obedience." ( "Letters of Aldous Huxley" published in 1969)
>Brave New World wound up being the more accurate picture of future society than 1984, despite being less well-known and referenced in cultural consciousness.
I encounter this exact thought in every comment thread that mentions 1984 without fail.
And usually 3 or so comments later I realise that they poster of this amazing idea hasnt read (or at least doesnt remember) 1984 anyway.
>Unfortunately, it seems like the former may be enabling the latter, so we may end up with a “porque no los dos” situation.
You mean some sort of situation where the masses (or "proles") are kept happy with puerile entertainment while those people with political impulses are kept under heavy surveillance? Kind of like the novel 1984?
Yep, there are the same tropes about 1984 in every discussion. And it's a shame, since 1984 presents something much stranger and more interesting than communism, surveillance dangers or dystopian universe in the abstract.
It shows what a system would evolve into without independent temporal continuity, and how truth can keep changing in a flow-like fashion while central power remains the only stable reference. Here, correspondence with reality is no longer treated as sacred, only useful. The Party does not need or want truth in any moral or objective sense. It only needs enough contact with reality to keep the machine running. Records still have to be managed, wars still have to be administered. Everything beyond that can be made fluid.
It also attacks the usual consolations: hope, love, that there's something human that remains untouchable. It shows how humanity is a fragile construct, and meticulously presents a procedure for how to break it
It shows precisely that no matter who you are or what you think you can do, a well-established system that seeks power for the sake of power will crush you just because it can. And if you're about to be crushed, they already declawed you systematically. Propaganda controls the present, manufactured records dissolved the past, Newspeak narrowed what you were able to think, manufactured dissent absorbed any idea of rebellion. And if the system wants you to love the act of being crushed, it has methods for that too. And why would it do that? Because power is an end in itself. And that's all you need to understand about unchecked power.
It also directly addresses that it's a new system and, unlike the Catholic Church during the Inquisition and unlike the communists, it is honest about its relentless accumulation of power, and doesn't need any kind of external legitimation from God or some placard of equality and prosperity for those who don't know better.
Let it also sink in that the book being ultra class-aware seemingly passes by almost everybody.
I’d love to know that as well. Long time Brave user that was curious about Vivaldi a long time ago and can’t remember why I went in the other direction. I want privacy and battery life on macOS
Firstly, it doesn’t seem to work for me and my wife - we hold the phones together but clicking start does nothing (and we’ve accepted Bluetooth etc).
Secondly, I wonder if you’ll have a massive chicken and egg issue with the physical feature. I get it’s the main feature but could you overcome it somehow initially while still maintaining your long term “gimmick”? Like could you allow people to connect with the first X friends (5? 10? 20? Whatever that can get virality and flywheel going) or connect with as many as you want virtually for the first X months etc. You could even have the contacts fade away slowly if they don’t get verified in person etc. You might want to model out different strategies (and be extremely conservative) otherwise you’ll be relying on lottery-level luck. Good luck anyway though :)
Yes. And it can be done in less "communist" ways; have countries' governments invest serious capital (even if they have to raise debt - they do anyway) in income producing assets related to AI, like large stakes in AI labs, building data centres etc.
Yes, doesn't need to be "communist" or even fully socialist.
I think governments should invest in their economies - mostly by investing in research, education, infrastructure, health and wellbeing of citizens, etc. but also putting capital into the later stages of expansion would make sense.
I certainly don't think people should not be able to start or own or profit from companies. But I do see a reason to limit their scale and/or make them more publicly owned beyond an certain scale.
I quite like the idea that "public" markets should become truly public, e.g. by some ratcheting percentage of public companies becoming owned by society at large over time (there would be several ways this could be done). This somewhat happens with the largest companies via index funds but only for those big enough to be in the indices and the distribution in unequal.
Maybe there are other/better ways, but it's pretty clear to me that big companies have a lot of negative impacts that aren't properly accounted for and so they are a very significant way in which a few people get richer at the expense of everyone else.
reply