hdevalence a day ago

It’s odd that the article identifies Apple’s hardware as a limitation for AI. I don’t think this is the case. If anything it’s the opposite, and makes Apple’s lack of execution more mysterious.

I was running Stable Diffusion on my iPhone two years ago. You can get quite good open weights models running on-device today. What’s going on over there?

  • tedivm a day ago

    I don't think there's much mystery here. Apple has blocked their researchers from publishing and has a very siloed approach. Researchers don't like this, so they work for companies that allow them to publish and engage in the research community. As a result Apple can't hire the talent needed to execute in this space.

    • Someone a day ago

      > Apple has blocked their researchers from publishing

      Counterexamples: https://machinelearning.apple.com/research/

      • tedivm 21 hours ago

        Sort by oldest. They had two papers in 2017 and three in 2018. It was only in 2019 that they started opening up more, but you can't change decades of culture quickly. The AI researchers I know (and I know a lot, having been VPE at multiple AI companies) would rather work with groups that are more open.

      • atonse a day ago

        They did start this but they were a few years late to the game, so they're probably having to play catch up.

        Also they still have a streak of secrecy so I wonder if any researchers are skittish thinking that Apple might just arbitrarily stop them from publishing certain papers?

    • 3cats-in-a-coat a day ago

      As if xAI publishes a lot. But even they have models that work.

      • tedivm 21 hours ago

        As far as I can tell xAI doesn't publish because they don't have anything to publish. It seems to me like they aren't doing anything novel or state of the art, they're just consuming other people's research and implementing it.

  • jerf a day ago

    I wonder if AI is just not up to Apple's standards yet. It's very, very amazing in some ways, but also, deeply flaky in others. Remember all the Apple news summarization memes? That sort of thing bothers any tech company but it really bothers Apple. "Very amazing but also deeply flaky" is fine for the Android ecosystem but it's not on brand for Apple.

    I suspect Steve Jobs would be very aggressively driving them internally but also not necessarily releasing much yet either.

    • WorldPeas a day ago

      This is much less of a Jobs Apple and more of a Sculley Apple

  • nstj a day ago

    > The problem there is twofold. One is that so-called on-device AI hasn’t yet proved to be a major selling point for products such as PCs and smartphones. The other is that Apple’s lack of its own cloud-based AI capabilities leaves the $3 trillion company still in need of powerful allies. The company struck a deal with OpenAI last year to effectively back up its own AI capabilities, and analysts have widely been expecting similar partnerships involving other major AI services, including Google’s Gemini.

  • hu3 a day ago

    I think the challenge is battery. Frequent inference draws a ton of power.

    • neepi a day ago

      Yes. If I have to trade off a stupid phone with 100% determinism and good battery life versus an intelligent phone with 70% determinism and shit battery life then I'm going to take the stupid phone.

      • postalrat a day ago

        I get 4 or so days of battery life. If I lost a day I don't think I'd notice.

        • socalgal2 a day ago

          My phone 15 pro is almost always dead by the end of the day. Yesterday it actually died and I had to plug it in. last night it was at 5%

          and, I’m not a major user. I don’t TikTok, X, Instagram, or any other social media. Nor do I do much else on my phone that I see heavy users do around me

          • conception a day ago

            You can see which apps are eating battery in settings. I do 80% charging doing about 6 hours of screen time a day and I usually end the day around 30-50%.

          • gessha a day ago

            Did you notice battery life tanking a few weeks ago or am I imagining things?

            • sitzkrieg 20 hours ago

              noticeable decrease after updating ios. yay

            • neepi a day ago

              15 pro here. I noticed it with 18.5

    • mingus88 a day ago

      That’s where PCC comes in. But I still don’t understand how their device margins are going to fund all these AI data centers

      Especially now that their App Store revenue is collapsing

    • spicybbq a day ago

      Meanwhile, the main attraction of the upcoming iphone 17 lineup is a thin and light "Air" model that has a bigger screen and a smaller battery.

      • sgerenser 19 hours ago

        Thinner with a smaller battery, yes. Not a larger screen though, actually slightly smaller than the "Plus" iPhone which it will supposedly be replacing.

    • ganoushoreilly a day ago

      I wouldn't say that's a Unique Apple Problem though.

  • roryirvine 21 hours ago

    Wrong or not, it's a danger for Apple when this sort of thing starts popping up in the media.

    It wouldn't take much for a "Apple have taken a wrong turn, and have left themselves unable to participate in the race for AI" narrative to take hold. If that happened, it would severely damage their position at the premium end of the market.

  • tmaly 17 hours ago

    All I know is that my iPhone 13 pro had better battery life than my iPhone 15 pro. Is this Apple Intelligence chip just draining my battery?

miladyincontrol a day ago

Its a low bar but a bar you cant walk back from if you collide poorly rather than pass with flying colours.

Take MS, sure they are doing a lot with AI, however some matters like Recall I'm not sure they can ever fully step back from, the trust has been lost for many by a rushed attempt.

  • xnx a day ago

    Recall-like features will be a standard and accepted part of operating systems soon. People freaked out about Gmail "reading all your email" at first too.

    I know there's a difference in the OS having access to everything on your screen, but people will still calm down once there are useful benefits delivered.

xanderlewis a day ago

Why would I want ‘AI’ on my iPhone anyway? All I want is an all-in-one electronic notepad, calculator, encyclopaedia, communicator, music player and camera. And it already does all of that (almost) perfectly. And has done for the last fifteen years.

  • mingus88 a day ago

    That’s it though. The iPhone set the paradigm for nearly 20 years. And up to that point nobody knew it was coming.

    It’s been done and every tech company wants to find the next game changing thing.

    I’m kinda rooting for Apple on this one. Give me a smarter smart phone that doesn’t sell out my privacy.

    If they can do that and whatever AR or VR or whatever ends up finally being usable perhaps it will be built with user privacy back in mind. As it is I’m not getting anywhere near a Meta or a Google or an Amazon reality device

    • xanderlewis 21 hours ago

      I agree mostly, but what exactly do you mean by privacy?

      I think the real pressing issue is not ‘privacy’ or ‘tracking’ or whatever; it’s having one’s views slowly but surely warped by AI-enabled mass disinformation campaigns and having one’s eyeballs taken hostage by whichever company pays enough to keep you watching. Half the world seems to now be addicted to short-form video slop/brainrot now, and they don’t even enjoy it!

      That’s really the hallmark of addiction — you’re consciously aware that it’s completely pointless and detrimental to your life and yet you can’t put it down.

      Perhaps the same was said of TV when it was new, but, looking at what the BBC used to broadcast as recently as twenty years ago, I’m shocked how far we’ve fallen. If Apple really wants to save the world, they should find a solution to this. Since (fairly uniquely) their business isn’t dependent on selling attention to advertisers, they might just have a shot.

      • mingus88 15 hours ago

        Privacy here means Apple is not a data broker. Your inference requests are completely anonymous.

        Apple made their money when they sold an expensive bit of hardware. They even use some of these large margins to build out private compute cloud to make sure they never see what you send them

        Frankly they are the only big tech company I will even remotely trust. The fact that their AI isn’t bleeding edge is a feature for me

    • sitzkrieg 20 hours ago

      apple's pro privacy stance is marketing

  • jimbokun a day ago

    That's a huge, near existential threat for Apple.

    If everyone's iPhones already do everything they want, the only reason to buy a new one is if the current one breaks. Which would decimate Apple's stock price.

    • neepi a day ago

      We're already there. I only buy a new one every 2 years because they last so damn long that my three kids can get them handed down that long.

      Just got rid of a 9 year old 6s a few months ago that still worked fine!

      • ryandrake a day ago

        This is the real tragedy: That we have to throw away perfectly working phones because their manufacturer and the software ecosystem are working together to make sure our devices outlive the software on them.

        • neepi a day ago

          The 6s still got an update earlier this year. I sold it to someone who was going to use it. It was just sitting around otherwise. So I'm not sure that's entirely true.

        • jimbokun a day ago

          Eh.

          Not sure you can fault Apple for stopping investment in software for devices past a certain age. And certain functionality just isn't supported by older hardware.

          Doesn't mean the devices stop working. Just that Apple is saying you're on your own.

          (Although the proprietary nature of Apple software means you also can't continue upgrading the software on your own or through a third party.)

    • xanderlewis 21 hours ago

      You’re not wrong, but how does it relate to my question? I’m asking why I would want AI, not why Apple would want to artificially inject ‘AI’ into their product — that’s obvious.

  • ryandrake a day ago

    I don't care if my products "have AI" in them.

    I care that they do the tasks I command them to do, quickly and efficiently. If the developer achieves this using AI somehow, great. If the developer achieves this using traditional algorithms, also great. It doesn't even remotely matter to me if they are "using AI" any more than it matters whether they are "using Python."

  • senko a day ago

    Want Siri to actually work?

    • whateveracct a day ago

      i have disabled siri for well over a decade now

      i don't need that sort of thing

      • msgodel a day ago

        I never liked Siri but leaving Perplexity/ChatGPT in voice mode so I can just verbally ask for web searches etc while I'm doing other things is amazing. Something like that with MCP would actually be revolutionary.

      • xanderlewis 21 hours ago

        Agreed. Even if it worked perfectly, I’d rather just press a couple of buttons than have to speak to it. It’s just of no interest to me.

  • giancarlostoro a day ago

    My biggest concern is that it will degrade my phone sooner and its battery life. I rather just hit an LLM in the cloud somewhere from my phone.

    If the tech ever reaches the point where Apple can fully do, on device-first and there's no issues with it, then they'll likely invest drastically into it.

    • codexy 12 hours ago

      Agree. AI features definitely eat up battery and processing power, even if they're not running 24/7.

xnx a day ago

Apple should power their AI features with Gemini, the same way they got their search feature from Google Search.

Yesterday's presentation shows that their software wheelhouse is making fancy user interfaces.

  • bastawhiz a day ago

    Hard disagree. Apple powers search with Google because Google pays them. Google pays them because they can show ads and scarf down user data. I don't want my devices' ability to do certain tasks controlled by the highest bidder for my attention and data.

    Gemini hasn't even been "the best" for a year. Google's quick answers AI is the most laughably bad high-exposure model out there. In six months, Anthropic or OpenAI or any other provider might have a model or models that are exceedingly good for the use cases that Apple cares about.

    Hell, Meta or Microsoft might surprise us and make a really robust model that uses very little in the way of resources and battery (arguably, MS has in the past and will do it again), and on-device compute will be the best option.

    There's no best option right now. And if you're not aiming for short term ROI on AI products, there's no reason to make commitments that you might regret in a quarter or two.

pi-err a day ago

Feels more like Apple is sticking to its position that "all those LLMs are gimmicks and not actual AI". So they think they're not late because they see the primary tech behind oAI (etc) as a base for non-threatening features.

This stance will probably age very poorly but that's what it is.

  • ryandrake a day ago

    They didn't do anything with "blockchain" either, and are no worse off for it. It's actually nice to see a company not just chasing whatever is hot and trendy each month.

  • rglover a day ago

    I'd say they're correct from the consumer POV. In the context of a phone that's consumed mostly by non-technical people, AI in its current form is at best a better Google that occasionally fabricates the truth (or a silly image generator for creating memes).

  • rstat1 a day ago

    >>all those LLMs are gimmicks and not actual AI

    If that's their stance, they aren't wrong.

ryandrake a day ago

Why is it just assumed that all companies should be "doing something with AI" and if they are not, then they are somehow "behind?"

This is like saying "Company X is behind on Python" when they just don't use Python to make their products.

  • conception a day ago

    When you onboard a 100 million users in a week, people assume there’s money to be made there.

    • ryandrake 21 hours ago

      No idea what this means. Who is onboarding 100 million users a week, and where is there money to be made?

neepi a day ago

Oh the press AI hit job. Here we go.

No one has reached the bar.

  • jimbokun a day ago

    How so?

    AI products have shown a lot of progress over the past few years, even if they haven't "reached the bar".

    But Apple is failing to deliver anything solid in the same time frame. Disregarding an arbitrary notion of "reaching the bar", they need to show progress.

    • mjburgess a day ago

      What has Microsoft delivered? What has any hardware manufacturer delivered? What features of any operating system have been added by any company, that are worth while?

      There's basically one "killer ai app", that's the chatbot. There's a second, plausible case, for agentic additions to IDEs.

      But where is this high bar we're meant to be seeing? All I see is google and stackoverflow being replaced by openai.

    • runako a day ago

      I think part of the problem is nobody has a clear vision of what "progress" would look like here. What is the thing that they should be shipping today that would get someone to upgrade their phone, or to switch to iPhone?

      Inverting it, what are the AI features that Google has shipped that are getting people to upgrade their Android phones, or to switch to Android?

      My Apple devices are pretty useful for AI devices now given that I can run e.g. ChatGPT and Claude on them. I know they have to do something so the tech press can stop saying they are behind, but it still feels premature given lack of consensus on how best to integrate phones with the unreliable AIs we have today.

    • BeFlatXIII a day ago

      > they need to show progress

      To whom?

      • jimbokun a day ago

        Customers deciding whether or not to buy new phone models.

    • neepi a day ago

      No one has delivered anything solid. Progress is asymptotic here and everything x months and y billion dollars away.

      We have shitty chatbots which confidently spew garbage and a new button on laptops no one wanted.

      And the customers are worried that the technology is coming for their jobs, their income and their lives.

  • an0malous a day ago

    That’s what I was thinking. The most valuable AI tools are probably coding assistants, and I’d argue they haven’t even reached the bar. What AI products have reached the bar?

lsy a day ago

On the one hand, I actually thought it was pretty refreshing to see a tech company keynote that wasn't wall-to-wall with just-so futuristic AI demos that will, if they ever arrive, inevitably be unreliable and finicky under real world usage. Apple has a pretty big opportunity to be a company that invests in its core product functionality and to add AI strategically where it makes sense, keeping its products performant and stable while everyone else sloppifies.

But I don't think they are consciously doing that. I assume like every other company they are throwing a lot of resources at the new hype and just failing, and this is the backup plan. I'm not sure what that says about the strategy at the top.

  • cosmic_cheese a day ago

    People keep talking about grand things like new paradigms, but given current limitations I wouldn’t be surprised if for the most part, the best AI winds up being that which you can’t see (the strategic application you mention). Think less miraculous everything machines and more specialized models enhancing things that already exist.

    The tendency for LLMs to flub and hallucinate gets downplayed but I think it’s actually the main blocker for it becoming a “smartphone killer” kind of thing. The problem is that fixing this issue is anything but easy and might even require actual reasoning capabilities (not the current Markov chain ping-pong “reasoning”), e.g. something adjacent to AGI, and I think we’re probably still many years away from that.

taco_emoji a day ago

Why do people think Apple is "failing" at AI instead of simply making the bet that AI, at least on a phone, is not something people really care about?

  • rglover a day ago

    Because they want to have their own belief—that AI is more important than others believe it to be—reinforced. They think it's important and significant, so anybody who doesn't must be an uninformed, unimaginative idiot (or a failing company).

    This will play out exactly how you think. The hubris will be confronted with harsh reality at some point, all of the chest puffers will get real quiet, and companies like Apple will look wise for not getting too involved, too soon.

msgodel a day ago

The problem with Apple's platforms is they can't choose to wait until they have it right like they did with IE vs Safari. The walled garden means their approach is the only one available to users.

If the mac worked the way the iPhone does the mac just wouldn't have a browser until after the .com boom was over. That's why I'm shorting them. Their inability to give people an escape hatch so their users can actually do useful things with what they build is killing their platforms.

  • WorldPeas a day ago

    I can't agree more. I hope a more convergent os (and hopefully an open and private one at that) is over the horizon, the problem being that as it stands, linux is lacks accessibility and the commonality factor that would make using it as a developer worth it.

    • msgodel a day ago

      VTEs are incredibly accessible and language models I think mean using languages will actually be the most accessible UI to non-technical users going forward.

      I'd argue GNU-style CLI and libinput based tools are actually more discoverable than GUIs now that smartphones killed CUA and friends. Is pressing tab twice really worse than "reverse three finger pinch?"

  • api a day ago

    I don't disagree but you're just talking about iOS. I run third party AI stuff on Mac including ollama and https://boltai.com which gives you a common interface for both local and remote models.

    iOS devices are just consumer devices for consuming content. They're almost a completely different platform.

    As for integrated AI, I kinda like that Apple is trying to get it "right" in the sense of being on device in many cases (or at least capable of it) and secure (privacy respecting).

    Their models are behind though. If they were really serious they would buy Anthropic or Mistral with all their cash. Anthropic might not sell but Mistral seems like something they could bulldoze with cash if they wanted, and they'd get better models and a better model-making team.

    • msgodel a day ago

      People said similar things about Blackberry near the peak. They're setting themselves up for disruption and decay.

    • 2OEH8eoCRo0 a day ago

      How much revenue does Mac bring in compared to iOS devices? Apple's success is entirely due to iOS/iPhone. The Mac is irrelevant

      • api a day ago

        More than a bean counter analysis would show. It makes them the dominant PC platform for developers and many forms of "serious work." Abandoning or ruining the Mac would, over time, relegate them to one of the "other" ecosystems from a developer POV. As everyone knows the "other" ecosystems are the ones that get less attention.

        • msgodel a day ago

          I do "serious work" from a mac and the only software I use on it is OpenSSH, Safari, Webex, and Outlook. Everything actually serious gets done on a Linux VM in a datacenter.

        • bigyabai 18 hours ago

          > It makes them the dominant PC platform for developers

          In my 12 years of employment in this industry, I have yet to encounter a MacOS build server. "Serious work" doesn't happen on MacOS by design; it's not a deployment platform. You don't put Mac instances on your K8s network, most of the time your customers aren't using a Mac either. Even big companies like 1Pass have realized what a waste this is, and started porting their Mac software to Electron. Apple couldn't dominate developers if they systematically murdered every other POSIX-compliant OS.

          The more time you spend as a development studio optimizing for a runtime you never use, the more capital you waste to achieve the same end-goal. It's not a real thing outside the comically capital-intensive SV culture.

joshstrange a day ago

I agree Apple is behind on AI, it's still up in the air if they "missed" AI yet (like MS missed mobile). I hope they are able to compete in the long run.

The Foundation models/SDK/APIs are a big step in the right direction but they are coming 6-12 months behind everyone else. AI is an area that it's not at all clear if "slow and steady wins the race" (Apple's normal MO). If there is a paradigm shift in how we interact with our phones using LLMs then Apple is going to get left behind as thing currently stand.

It's also damning that Sky [0] didn't come out of Apple. More damning is that the majority of the team behind Sky _DID_ come from Apple (and Workflow, aka Shortcuts) before that. Why did this team feel they had to leave Apple to create something like Sky? Maybe it's a ploy to get acquired again but I think that Apple just isn't good at LLM-based products because of the uncertainty. Apple wants to avoid the "glue on pizza" or "stop taking your meds" moment but it's causing them to miss the boat completely.

LLMs are far from perfect, they lie/hallucinate or just lose the plot. That said, there are some really valuable things you can do with them. Historically Apple has been unwilling to accept the downsides and would rather wait to ship but with LLMs I don't think they are going to ever get a "no downsides" approach to work.

[0] https://www.macrumors.com/2025/05/28/sky-ai-mac-app/

  • runako a day ago

    > it's still up in the air if they "missed" AI yet (like MS missed mobile)

    I think the clock on this won't start until someone ships something compelling that is not available on Apple devices. Currently, the leading AI tools have first-party iOS apps.

    > Why did this team feel they had to leave Apple to create something like Sky?

    No personal knowledge of that team, but the economics around an acquisition of an AI-based product in this market are very different from the economics around "generous bonus for shipping a successful product."

  • jimbokun a day ago

    > It's also damning that Sky [0] didn't come out of Apple. More damning is that the majority of the team behind Sky _DID_ come from Apple (and Workflow, aka Shortcuts) before that.

    Time for an acqui-hire?

codingwagie a day ago

Real reason is apple is probably stacked with legacy employees with 30 years of experience in some hyper specific apple stuff, and now with the ai paradigm shift, legacy employee expertise is not only obsolete, but is probably a hinderance. They will recover, will just be slow