Let Tim Cook

They did not even check the stocktic anyway. The presentation given at Apple's Worldwide Developers Conference, usually referred to as WWDC and where Apple usually announces its next software updates, felt remarkably demure in comparison to May's Google IO conference, which is Google's equivalent, where CEO Sandhar Pashai and Head of Search Liz Read hype the next generation of Google

products and search updates that absolutely nobody asked for. Historically, what WHDe Developers Conference is where Apple announced major updates

to both its product and software lineups. A WWDC two thousand and five, the late Steve Jobs announced the company would move from IBM designed and Motorola manufactured power PC processors, allowing for faster and more power efficient computers, as well as the ability to duel boot Windows, a huge moment for Max that was met with rancorous applause from people

that really should have known better. Fifteen years later, Tim Cook, now CEO, would announce Apple's shift away from Intel stagnating X eight six architecture to Apple's homegrown ARM based processes, starting with the M series processes that have now become ubiquitous across MAX and iPads, one which was met with the bombastic promise that these were the most powerful chips ever created. I should add, by the way, the M series processes are actually probably one of the better things

that's come out in the last ten years. The average MacBook Air is insanely fast now, like I have a MacBook Pro. It's phenomenal. A lot of you say, I'm not positive, Okay, these are things I like. Here are things I like. Are you happy now? I am kind of anyway. WWDC is where Apple has unveiled every new version of the Mac operating System OSX. It's where the original Mac Pro, FaceTime and the iPhone app Store were

introduced to the world. This is the conference where Apple flexes its muscles and boasts about how powerful and important Apple is. Now you can understand why this year was just so strange. Gone was Apple's trademark bombast, It's propensity to tell the world how good it was at everything and how important its next big thing was. Instead, we got this weird juxtaposition between we're putting AI on your phone and it's totally fine. It's not big deal. You

shouldn't be scared at doors. It's not that big a deal.

Please don't be mad at us. Despite this ostensibly being Apple's big artificial intelligence play, CEO Tim Cook and SVP of Engineering Craig VEDERIGGI spend far more time explaining how much they love privacy and shareish your data and would never share it with anyone, and listen to all the ways we're keeping it private, before casually sauntering into a series of product updates that appear to actually use AI for a purpose and kind of useful and hardly suggests

that Apple is intentionally trying to distance itself from the

AI hype train. And while it's integrating CHATGBT intoosaighteen, which don't worry, I'll get to later, it's doing so an arms length approach, with most of the day to DAAI work done by the company's own models and actually on your device these AI integrations, which arrive some time in the fall to the iPhone fifteen Pro and iPhone fifteen Promax that's autumn for my British readers and listeners, along with iPad and MacBooks with M one or later chips.

They're fairly straightforward and don't do anything we haven't really seen elsewhere, which is kind of the Apple game. Apple's AI can generate transcripts of cause, but not before warning participants that it's doing so, generate images through a tool called image Playground, help write and rewrite emails kind of like grammarly does, and perform distinct actions across multiple apps, like adding slides to a presentation or opening a web image with the photos app where you can edit it.

The last items a little bit fuzzy is it extends to both Apple's own apps and other third party apps, and the functionality itself is expected to roll out over the coming year or so, and as ever, take any promises made without a demo with a big grain of salt, which I guess is just a regular grain of salt when you think about it. Apple also says that Siri will soon have an awareness of what's on your screen

and be able to respond appropriately. In one of their demos, you're filling out a form that asks for your driver's license number. Siri can then look through your photo library, identify any pictures of your license, and grab the pertinent information for you. Equal parts useful and creepy and also I really do not recommend having a picture of your

driver's license saved on your phone. But thanks to the power of artificial intelligence, you can now generate your own custom emojis, a feature that screams, we have to fill a few minutes of this goddamn presentation, But nevertheless they were quite excited, more excited than the open Ai integration, which, as I've said, I will get to you later as it requires a bit of space for me to laugh.

Apple claims that many of these features run entirely on your device, and those that don't use something called private cloud compute, where Apple claims in a lengthy privacy statement I'll link to that at no time will anyone ever have any access to any of the data that's being processed on their servers, even and I quote during active processing,

meaning when Apple's servers are handling the request. Apple has also offered security researchers the opportunity to personally verify the claims it's making, and in general the news it seems to have been well received, though some, like Chief Privacy Officer Steve Wilson of EXABEAM told Dark Reading, worry that threats in generative AI are poorly understood and that despite Apple's best efforts and no doubt heavy policing, some will

slip through the cracks. On some level, Wilson is being a little alarmist, with his only criticism being that we don't know everything about this thing generative AI, and thus should be scared about what we don't know, which isn't wrong. On the other hand, he's right in so far, and that we are relatively early days into the mass scale use of large language models like Apples, like GPT, like

anthropics Claude. Apple has chosen, albeit cautiously, to take the risk of integrating these models into devices at a time when companies like Google and Meta have proven that one cannot simply trust that a multi trillion dollar company will maintain the stability and privacy of a product, or for

that matter, can guarantee the accuracy of its outputs. Also really doesn't help that Apple CEO Tim Cook told The Washington Post that Apple's AI is short of one hundred percent when it comes to hallucinations, which, as a reminder, are when these llms tell you something authoritatively that isn't true. And this is a little bit worrying when we're talking about an AI model that's taking actions across your apps.

These features are predominantly running on Apple's own models, like I've said, so, not on GPT or claud or anything, which, according to Apple, are trained using licensed data such as images licensed from Shutterstock, data selected to enhance specific features.

No idea what this means, but it's from licensed data at least, as well as publicly available data collected by Apple's WebCrawler Applebot, which Apple allows publishers to alpt out of using a rule added to their website's code, which I should add, does not solve the bloody problem if Apple has already used the data for training purposes, which is bordering on impossible to confirm, I should add, I have added this code to my website. You're not getting

my blogs, mister Cook. And the thing is, if Apple already has done this, if they've already scraped this stuff, if they've used publicly available, which could mean any website on Google, for example. It's indicative of some really shady, shitty, underhanded behavior. Well, it's fair to say that Applebot the webscraper,

isn't you. It arrived at some point in twenty fifteen, it wasn't really publicized that much, save for a few posts on Apple rumor sites speculating that the company was working on a replacement to Google Search, which by the way, had never happened because Google pays over ten billion dollars

a year to Apple for that monopoly Jesus Christ. Prior to the launch of Apple's generative AI features, Applebot's sole stated purpose was to provide data for Siri and Searchlight, the search built into Mac OSX that scours both the user's hard drive so that it can pull data out so say you're looking for a file, or it can search for something on the Internet. Generally a useful feature, but an entirely different one to a generative AI one.

If Apple is already training its models on data scraped by Applebot, it will be immensely deceptive, using a tool way beyond its intended purpose, or at least its advertised one. If this came to pass, those unwilling to have their content repurpose this training data for a trillion dollar take company would really have nowhere to go. What are you going to do? Getting an AI to unlearn something isn't exactly straightforward, and Apple has yet to explain how that

might work. I've reached out to Apple for comment here, but it's unclear exactly how it intends to fulfill this promise that you can opt out of being included in their training data, especially if the applebot crawler has already allowed said data to be fed into the model. If I hear back, it'll be a little bit of a pain in the ass, but I will update that here. And because I love you all, I love you all so much. Though Apple has done a better job than most,

it's gross. It's disgusting, especially for the generally the best of the big tech companies to do the thing that all these people have done. It's disgraceful that, yeah, another big tech company has seen the open Internet as its personal property, despite publishing a remarkable amount of information about

their privacy standards and how their models are trained. Apple likely hopes that this privacy focused media blitz will hide the fact that it's ripping off everybody's work, just like Google, just like open ai, and just like Anthropic and Meta, except it's doing so in a way that's a little bit easier for the media to swallow, because the main selling point of its AI is not vomiting out oodles

of anodyne business bullshare. Unless you're using the smart reply feature in mail, which drafts a response to an email for you. I'm just I really don't understand people need that. I'm sorry if you can't email someone back with your brain words, what's going on? Seriously, what's going on? But it sucks, though, It really sucks even for the smallest feature to have to steal people's work, especially when you

have what two hundred billion dollars in the bank. Just a complete mark of disrespect against anyone who isn't an Apple user, and plenty of people who are. This is particularly worrying when you consider Apple's image generation features, which will function similarly to chat GPT and stable diffusion and yes, I'm getting to chat GPT. Well, I'd love to believe that Apple has only used licensed content to train its models.

It really isn't the case. And if their version of publicly available includes things from Google or DeviantArt or social networks. Apples just part the problem. They just part the pillaging of the Internet. They are stealing, just like all the other big tech companies, and there's just this is a time when I feel a little helpless forever. I don't know what you do here, and I don't think we're

going to hear anything back from them. But as I mentioned, I've requested comment from Apple about this, and I hope that I get an answer. Apple's approach to PR doesn't film me full of hope, though. It's almost as contemptuous as testless. They only really provide comment and interviews or review samples that those journalists they've deemed worthy, and only

really on certain subjects. It's a style that mirrors the kind of haughty, snooty image it crafted in the mid two thousands with those goddamn I'm a mac ads, with reporters divided into worthy and unworthy camps, meaning, by the way, those who would bend the knee to mister Jobs and his company and those who would not. And by the way, there weren't many who wouldn't. But okay, if you can put these very real concerns aside, Apple's AI announcement feels

equal parts useful and extremely strange. These are not world changing integrations. They're useful ones transcriptions, a better SII that works like it's meant to, that can take actions across multiple apps and know what you're talking about based on what you're doing on the screen theoretically at least, a better photos app with better search, which is already in the product already, and the ability to edit photos with AI and remove things and backgrounds. These are cool features.

These are the kind of things that Google would staple onto a pixel phone to pretend you're going to buy one instead of an iPhone. And these are all things that a user could foreseeably want to use in their daily lives without being told that they're in the future,

in large part because they are not. In fact, I'd argue that Apple's biggest generative AI push is in trying to sell us back the idea that Siri can actually do stuff, years after most users accepted that it was kind of a voice controlled roulette wheel that occasionally understood you, and for my Scottish, British and other people with accent friends,

that's very occasionally. What's weird, though, is my Apple TV remote can understand me every time I say like the Imaginarium of Doctor Parnassis, and it will pop up immediately. But I tell my phone, hey, can you play this song on this home pod? And it just freaks out. It has a complete conniption. It's so strange. Nevertheless, putting all my jokes aside, looking at everything here, I just

can't get over how reserved Apple is being about artificial intelligence. Well, Google has desperately tried and completely failed to convince Wall Street that it's building and selling the future by putting

generative AI into search and making it worse. Apple's almost desperate to explain how boring and normal your iPhone and Matt could be, and how Apple intelligence is just another feature that will make you want to keep using Apple products, rather than the reason that Apple should be worth forty trillion dollars and keep growing forever. I like how wireds Will Knight put it. He said that AI is a feature rather than the product. And this is what's really

interesting about this. Apple really doesn't feel like they're selling us the future. They feel like they're selling us the present. And maybe that's a good thing. Maybe it's good that Apple isn't rushing, but also the fact that Apple isn't rushing kind of says something about the wider AI movement. It suggests that outside of what they're offering, they don't really see much value to it. And now we reach

the funny part. Quietly stapled onto the end of the announcement of Apple Intelligence was integration of open AI's chat GPT in arguably the vaguest least consequential way that I've

ever seen a product launch. After spending an hour and a half, actually more so talking about how great its own AI features were, Craig Federighi mentioned that and I quote, there are other artificial intelligence tools available that can be useful for tasks that draw on board world knowledge or offer specialized domain expertise just fuck just completely stupid anyway, saying that Apple wanted users to be able to and again I quote, use these external models without having to

jump between different tools, integrating them directly into Apple's OS. Starting with the pioneer and market leader, open Ai. Apple's integration with open ai, the supposed big dog, the big swinging dick of the tech industry, in the biggest announcement for the company in years, was explicitly advertised as the first of multiple integrations with multiple models, and not even kidding literally that was the end of the announcement. It's

very funny. And even then, chat GPT's integration with Siri is entirely opt in, with certain requests that they were very unspecific about, occasionally prompting Apple to ask you if you'd like to run them through chat gpt, like I have these ingredients, what meal can I make? And other questions that tens, maybe even hundreds of people will find useful.

Or having chat gpt write a bedtime story for your kid if for some reason you lack books or creativity, or you feed into this weird thing with AI people where they think everything has to be about you and that you must be the star of every story. That's a separate podcast ed put it away. Chat gpt can also generate images and summarize documents, features already available in

Apple's own AI. This underwhelming addition to an already placid announcement ends with Apple adding that it intends to add support for other AI models in the future, before moving

on to how developers can integrate Apple's own AI. The deal that would supposedly cement Sam Mortman's hold on Silicon Valley and Open Ai ended up being a two minute long sidebar at the end of a near two hour long announcement, one that requires users to agree every single time they interact with it, with no specifics about how

often that a user would actually see chat GPT. Despite the insane romanticization of Sam Wrtman's incredible technology, Apple's approaches one that begins and ends with them saying do you really want to use this? And a remarkable lack of excitement or trust While it remains to be seen exactly

how often you'll be prompted to use it. Chat GPT's addition to Apple products feels far more like something cooked up to police Wall Street, and it worked, briefly, making Apple the most valuable US company, beating Microsoft just for a minute, or without having to invest billions of dollars into another company. Now, quick side note, there's undoubtedly here a level of cya covering one's ass here. If Apple could, it would undoubtedly monopolize AI on the iPhone, much like

it has without distribution and browser rendering engines. That's the bit of the web browser that turns HTMLCSS and JavaScript into a web page. From the beginning, Apple has saw absolute control of the iOS and its derivatives ipied os, tvOS and watch os ecosystems, using user safety as a

justification for well, they're kind of benevolent dictatorship. The problem is the line between protecting users and anti competitive behavior is thin, and indeed, regulators, particularly those in the European Commission, aren't convinced that Apple's iinclad control of the iPhone is entirely altruistic. And that's why you can access alternative app stores in the European Union and even install apps directly

from the web. And it's why you're no longer tied into using the Safari WebKit rendering engine, even if you install an alternative browser like Brave or Safari or Firefox. Indeed, if Apple iced out open AI or Anthropic, they both could conceivably complain about Apple to the European Commission, who might rule in favor of them and order Apple to start supporting their models, in addition to levying a huge

multi billion dollar EUROFIE. And so Apple chose to save itself the hassle and the cost and buried open Ai and the rest of them in the back. One might imagine though, putting aside by cynicism, this is still a big deal, right, Wow, Chat GPT on every iPhone, well, iPhone fifteen pro and so on. That's millions, hundreds of millions of people, tons of M one and M two and M three mags. It's a huge deal. Each time that someone interacts with chat gpt. Even though they're not profitable,

that's still a sliver of revenue. They're still paying chat gpt, right. They're not just stapling this on, Sam Altman isn't doing this for exposure? Is he? Yes? He is. This is not a coup at all, at least not one for open Ai. As reported by Bloomberg's Mark German, Apple didn't give open Ai any money to integrate Chat gpt, paying them in distribution. As I said, for a tool that loses the money on literally every transaction, Apple doesn't appear

to be paying anything. German also reports that users will be able to upgrade their chat GPT accounts to chat gpt Plus through the integration. No idea, I'm sorry, it's far you get more requests, I get, who cares? But the important thing is that upgrading to chat gpt plus will likely operate on the same terms as every other digital good and iOS, meaning that open ai will be paying thirty percent of that take to Apple unless the

user upgrades directly on chat GPT's website. This is not good for this company, to be clear, In most cases, companies integrating chat gpt pay them on a per thousand token basis, meaning that open ai will, while unprofitable, still generate revenue of some kind when people ask them stuff. Yet Apple's deal doesn't appear to pay them at all, meaning that every single time that someone uses chat gpt

on their iOS device, they will be bleeding money. The more popular or it is, the more money it's going to lose them. It's a terrible deal, folks. This deal is equal parts perilous and hilarious. It's so funny, and it's something that could genuinely end up hurting open Ai,

all while insulating Apple. In the event that this integration actually sees adoption by Apples hundreds of millions of users, It's going to cost open ai incredible amounts of money thanks to the fact that generative AI is both compute and energy intensive unlike any other piece of tech, especially as I imagine most users will ask it the occasional question and find no reason to opt into the twenty dollars a month Chat GPT plus subscription. Why would you care?

What is the use of this? On some level, it also shows a remarkable amount of disdain from Apple towards open Ai. Most deals that hinge on the vast reach of the iOS ecosystem involve some kind of exchange of money. In twenty twenty two, Google paid roughly twenty billion dollars to be the default search engine on iOS. And just to be clear, that deal very clearly a mob style thing. That's just Google saying, look it would be won't please don't build a search and please don't please don't build

a search engine. Our search engine's really bad. Just slim me, let me pay you man. Really gross stuff. But given the cost of running chat GPT, and the fact a large language model query is inherently more sophisticated and computationally expensive than a simple search one, you'd expect the opposite to be true. Right, the Apple would pay open ai something to defray its costs, but no, no money's changed hands.

And if open ai makes a sale directly on the iPhone, it'll undoubted least be subject to the same Apple text as every other company. Like I said, this is possibly one of the worst deals I've seen in my life.

It's a terrible deal, of folks. And I'd also speculate that Apple likely requires some level of service level agreement for its partners, meaning that open ai has to dedicate resources to maintain up time for Apple devices using chat GPT for free, which will in turn be incredibly expensive to maintain, burning money with every query on a product that Apple will only offer when Apple services can't do the job, and if iOS eighteen doesn't bring the expected

users to chat GPT, those resources are just gonna kind of sit there, unused, redundant, and they could be servicing other deeply unprofitable demand. It might be hyperbolic to describe this as text equivalent of the Versie tree, but only a little bit. It's also just unbelievably funny. It's so funny.

You heard the last episodes. You heard people talking about Sam Altman, like the art of the deal guy, the deal master, the master of contracts, the guy who can outthink us all, the mega genius, the mega nerd, the ultimate deal maker who could split the valley asunder with a wave of his hand. Sam Ortman is regularly described as this tier one operator, this superior intellect, and yet he has signed the worst tech deal I've seen other than I think the one signed by Yahoo with being

from a few episodes back. It might actually be worse though, because even then Yahoo got paid by Microsoft. Jesus Christ. I think this also proves something else open ai. They have no leverage, They don't have anything they can play with. They don't have anything they can go to Apple and say you need us, baby, No, they haven't got shit. It's hilarious. Open ai has effectively agreed to give chat GPT for free to hundreds of millions of people and

got absolutely nothing in return. And worse still in this marvelous technological cocultry. According to Anissa Gardisi of The Information, Apple is already working on cutting deals with both Google and their Gemini LLM and Anthropics Claude to integrate them into iOS, likely using open Aiy's dogshit deal to leverage better terms. Allmand may have been at the Worldwide Developers Conference.

He was there, people taking pictures of him acting like it was this big goddamn deal, but he didn't get to speak, nor was he mentioned during any of Tim Cook or Craig Federiggi's remarks, Apple treated chat GPT with less excitement than a new suite of emojis, kind of like an afterthought to a presentation that was deliberate and

intentional in its lack of froth or hype. A series of fun, potentially exciting features, things you could play with if you want it, all delivered with a continual promise of privacy and reliability that tacitly accepted that the AI hype train was moving a little too fast. And it's a humiliating moment for the generative AI movement, especially so for open Ai, a company that desperately needs good news at a time when the world has become deeply suspicious

of its product and its CEO. Apple is, for better or or worse, the gatekeeper to the technology used by hundreds of millions of people, and Apple has decided that chat GPT is a feature, not a product, and one that isn't trustworthy or useful enough to run without a user's permission or even pay for. This was, on some

level open AI's iPhone moment, though. This was the time the world would see exactly how important the most important tech company in the world thought the most important startup in the world was and the answer was not particularly

important at all. Apple can and likely will dump chat GPT at a moment's notice, replacing it with any number of other large language models, or ripping the feature out entirely thanks to the fact that it was introduced as a feature at the end of another feature, at the end of a presentation of better, more useful features. And what's insane is all of this is done without you creating an account. I know some of you might be a little bit scared that chat gbt's in your phone.

It isn't. They don't get your queries, they don't save anything about what you're doing on their Apple has raked them over the coals, and you don't even need an account to use it, which means that they're not even increasing their users with this. Their queries are going up, but they can't even get any vanity metrics out of this. It's so unbelievably busted. It is one of the funniest and worst deals I've ever seen. I think it also

shows that Sam Mortman doesn't have the juice. He really hasn't been able to outside of kind of intimidating people vaguely in the valley, really use his power that much. Microsoft weighed in to get him back at open Ai because it was good for Satchy Nadella, who had kind of tied his fate to Sam's. And it's funny the moment they step into the real world just doesn't seem to scale that well. Where are the big deals for

open Ai? Where are the huge integrations? The Information reported last week that open AI's annualized revenue, which by the way, is just a fudged figure. It takes the months of revenue so far and extrapolates from there. They say they're annualized revenue is going to reach the billions in twenty twenty four. Yet they're still so unprofitable. And even if Apple had paid the money to integrate Chat GBT, it would still lose money on every search. And it's so bad.

I'm not sure why no one else is saying this, and maybe some are Nick Builton and Vanity Fair give you cred there, classic dagnam lad. And the thing is the few people saying it with me are also smart. I don't know, but this is a crazy bad deal. This is one the guarantee is that open ai will bleed money every single time someone picks up if they even use it. If they don't use it, that's still embarrassing.

It's still a deeply damaging thing for open Ai. Sam Altman, despite being the Valley's golden boy, didn't get to go on stage. They didn't even in this presentation mention what chat gbt's premium features do, nor did they really make much effort to explain why you should give a shit. And I'm not trying to romanticize Apple. There are no great, big tech companies all out my Apple devices. I'm looking forward to some of these features. That is faint praise

for a tech industry that's lost its way. But I will give them credit for tim cook cooking mister Altman, because this is a great time for people who like bad things happening to bad people. I know I shouldn't be calling Sam Mormon a bad person. I mean, just read the stuff about him, listen to the episode. But look what happens when a little boy from the world of startups tries and goes up against the ugly forces of capitalism, the real scary monopolies. Apple won't accept another monopoly.

This is the empire, for better or for worse. It is scary how much power Apple has. It is worth considering whether you are feeling safe using AI on these devices. What they've said suggests that, yeah, I would say you are, but I think that's something you really should consider. And in the episode links, I will provide all of the documentation of everything I'm talking about here, because I want you to genuinely know everything you can make an informed decision.

I will say I would be unsurprised if every single other phone company doesn't do similar stuff. Samsung's already doing AI stuff I haven't looked, but I amaguine Google is too. Everyone's going to be doing this. I would trust Apple above them, because ah they they apparently have clearly worked out that what people actually want is for their phones to work, and that they'll keep making money just by making people happy. I'm not going to say they're super ethical.

They're better than most, and they say that with a great deal of kind of annoyance because they don't really want to blow smoke with them. But back to the really fun part. This may be a genuinely dangerous thing for open AI, and at the very least, it's deeply horribly humiliating. Sam Altman might be able to walk into any room in the valley and say insane things like I need seven trillion dollars for new energy. I needed one hundred billion dollar computer, mister Nadella, But when he

walks up to Tim Cook, there's nothing to really say. Oh, yeah, we'll add this at the end so that Wall Street feels happy, and that's all this is. This is so idiots like Jim Kramer on CNBC can say, Wow, Apple's god chat GPT in it and great, it doesn't matter, it's buried, and I guarantee you you're gonna be able to turn it on. Apple aren't the best. I don't know who the best are. I don't know who the most ethical are. I don't even know how you evalue at that. You know what, Let Tim Cook thank you

for listening to Better Offline. The editor and composer of the Better Offline theme song is Matasowski. You can check out more of his music and audio projects at Matasowski dot com, M A T T O S O W s ki dot com. You can email me at easy at Better Offline dot com, or visit Better offline dot com to find more podcast links and of course my newsletter. I also really recommend you go to chat dot Where's Youreed dot at to visit the discord, and go to

our slash Better Offline to check out our reddit. Thank you so much for the listening. Better Offline is a production of cool Zone Media. For more from cool Zone Media, visit our website coolzonmedia dot com, or check us out on the iHeartRadio app, Apple Podcasts, or wherever you get your podcasts.