Everyday AI Podcast – An AI and ChatGPT Podcast - EP 63: Apple GPT - Is Apple coming for ChatGPT?

Episode Date: July 21, 2023

Apple has finally built its own GPT! Now that Apple has arrived to the AI scene, what does that mean for the rest of AI and Apple's competitors? Are we seeing the rise of ChatGPT's rival? Today we take a deeper look.Oh. We also covered some breaking news on the US government’s AI agreement with major AI tech companies.Newsletter: Sign-up for our free daily newsletterMore on this: Episode PageMore on this topic in today's newsletterJoin the discussion: Ask Jordan questions about AI and Apple's GPTUpcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: [e-mail] with Jordan on LinkedInTime Stamps:[00:1:14] Breaking News: AI companies and US government reach agreement - breakdown and discussion[00:11:50] Breakdown of Apple's new GPT[00:13:10] Apple adding GPT to Siri?[00:15:34] Will Apple's GPT be better than the competition?[00:20:10] AI is driven by monetization[00:23:11] Final RecapTopics Covered in This Episode:1. Introduction: Apple's Entry into the Chat GPT Market- Apple's plans to build its own chat GPT system- Speculation on the potential success of Apple's GPT system2. Voluntary Agreement Between Major AI Tech Companies and the Government- Details of the agreement- Areas of focus: transparency, content watermarking, bias and discrimination, privacy, third-party testing3. Concerns and Open-ended Questions about the Agreement- Speaker's reservations about self-regulation by big tech companies- Questioning the effectiveness and potential loopholes of the agreementKeywords:Apple, Apple GPT, ChatGPT, US AI agreement, AI tech companies, US government, transparency, content watermarking, avoiding bias, discrimination, protecting privacy, third-party testing, concerns, open-ended questions, big tech companies, self-regulation, Microsoft, Google, generative AISend Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info) Start Here ▶️Not sure where to start when it comes to AI? Start with our Start Here Series. You can listen to the first drop -- Episode 691 -- or get free access to our Inner Cricle community and all episodes: StartHereSeries.com Also, here's a link to the entire series on a Spotify playlist. 

This is the Everyday AI Show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business, and everyday life. Meet Firefly AI Assistant, now live in Adobe Firefly, the All In One Creative AI Studio. Just describe what you want to create and the assistant handles the rest, orchestrating multi-step workflows across Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome. The assistant accelerates execution. Apple is building its own chat GPT rival.

That's right. Last but not least to the party, Apple has built its own GPT system. We're going to talk about what that means, and I'm going to give you two billion reasons why it'll probably be more successful than you think. So thank you for joining me. My name is Jordan Wilson. and I'm the host of Everyday AI. This is your live, daily, live stream, podcast, free daily newsletter,

helping everyday people like you and me, keep up with what's going on in the world of artificial intelligence, because there's a lot. And how you can actually understand it, right? That's the most important thing. That's what we're here trying to do. So if you are joining us live, please drop a comment. Let me know what you think about Apple's new GPT,

But even more importantly than that, we got some breaking news, right? So we always start the show with news, you know, giving you the top, you know, three, four headlines that are, you know, going to affect you. But we're going to do a little bit, we're going to do something else today. We're actually going to talk about one big news piece. So let's talk about that right now. And yes, good morning, Dr. Harvey Castro. Thank you for joining us.

So, yeah, let's, before we talk about Apple, GPT, because I have a lot of thoughts on that. There are some literal late-breaking news. You know, when I went to bed, this wasn't a thing. I woke up and it happened about an hour ago. But there has been some initial agreements between the major AI tech companies in the U.S. And the government. So let's talk about what that means. Thank you for joining us, Nadia, hello, from Ann Arbor. As a reminder, if you're joining the live together, let's talk together. Leave me a comment.

But if you are listening on the podcast, click in the show notes. You know, you can get a link to go check out more of the episode, but also a link to join us in the future here live on the show. All right. Yes. Good morning to Rastafa as well. So let's actually talk about this breaking news. Don't worry. We're going to get back to Apple GBT because I think that's important.

But this was just, just released. So here's what this new voluntary agreement is and what it means. So as a reminder, as of at least now, this is just a voluntary agreement between the White House. So President Biden is supposed to be meeting today. Today is Friday. So he's supposed to be meeting with leaders of seven big companies. So the CEOs or presidents of these companies.

Google, Microsoft, Meta, Amazon, OpenAI, Anthropic, and Infliction. So those are all, you know, it's kind of funny, you know, Apple's not on that list. So what's the difference between all of these companies in Apple? Well, all of these companies right now, the seven that I just named, Google, Microsoft, meta, Amazon, OpenAI, Anthropic, and Infliction. they all have, I believe, except Amazon now that I'm looking at this, but the other six all have publicly available AI chat. So you have, you know, your Bard, your Bing chat, you know, Facebook just released their Lama 2, which is going to be kind of their kind of quote-unquote answer

to open AI. And then you have OpenAI, Anthropic Cloud, an inflection pie, which we talked about a little on the show yesterday. So what does this mean? and let me know what you guys think of this breaking news of AI companies and the government reaching a tentative agreement. A couple more details on this, right? So what is the agreement? It's number one, like I said, it's voluntary. It's going to be kind of released today. This is just breaking on a couple new sites. But here is the information that we have so far on what this voluntary agreement is, what it means, and also how it's going to affect all of us. So the lead thing is it's based in transparency. So they want all of these companies to be more transparent with the data.

They want them to be more transparent with the government and more transparent with you and me to say, hey, here's where we're getting the data, here's how we're using it when you tell us something, all of those things. Yes, and I agree, Nadia, with your comment, Lama 2, having an open license is a big move from that. Absolutely. We talked about that on the show. So, yes, meta is releasing in partnership with Microsoft, Lama 2. So kind of their version of GPTS open source, extremely important. So let's get back to this new agreement between Big Tech and the government. So transparency with the data, how it's being used, all of that.

Also, to denote when content is generated through a watermark. That piece is interesting. So it's obviously easier said than done with. images, right? And Google has talked about that in their bar chat. But, you know, there aren't details on this yet, but I don't see how content will be watermarked, written content will be watermarked, right? So maybe it'll spit it out as part of the, as part of the response or as part of the output, you know, like, hey, this was generated with AI. But I'm guessing content creators or anyone else is not going to include that little piece.

You know, it's like by not including it, right? Because some people will, but if you leave it out, all of a sudden you are saying like, oh, look at all this great content I'm producing by myself, right? So the advantage or the pros of including this, if it is just a written watermark, it doesn't make sense to me because most people are not going to include it. The other kind of in this pledge, so to speak, right, because this isn't legislation. You know, this isn't a law. This is an agreement or this is a voluntary pledge by these companies that they're going

to abide kind of by these rules. So we talked about information sharing, data privacy, watermarks, but also they said to avoid bias and discrimination. That one stuck out to me because it's very hard to do. to do. It's very hard to do because, you know, even Open AI, so the biggest company out of all these in terms of users for their large language model, you know, Sam Altman, the CEO has essentially admitted that there is bias in the models because they are trained by humans. And all humans implicitly have bias and discrimination. So that is worked in there regardless of what you think

you're not. So that piece is interesting. Also, there's a pledge, quote unquote, to protect privacy. And here's the other one, here's the other big one, is they are going to submit to all of their systems being tested by a third party before being released. All right. That one's interesting because what you essentially have here is you have self-governance, right? And I can see the concerns that people might have with essentially big tech and AI regulating itself. But if I'm being honest, it kind of does make sense. because it is the best way to, you know, and some of the smartest people in the world have said this.

You know, Bill Gates said this recently. The best way to regulate AI is with AI, right? So should these big tech AI companies be regulating and policing themselves? Probably, I don't know, maybe, maybe not. What do you guys think? You know, there's so many big concerns and open-ended questions with, you know, these seven companies, and I'm sure there's going to be more. I'm sure Apple will be thrown into the mix when they release their Apple GPT, which we are going to talk about here.

But there's a lot to unwrap in this story. There's going to be more details being released over the next day after they officially have this meeting. This is just some initial kind of pre-meeting reporting. And maybe even by the time that you're listening to this, you know, if you're listening to this on your afternoon drive home, If you're here in the U.S., a lot of this may have already been settled. But let's quickly take a question or two. And then we are going to get into Apple GBT, because I think it's important to talk about.

And it's exciting as well. So let's take a look. We have a couple comments. Yeah, sorry. It looks like there's an odd start to the show this morning. Yeah, live stream. Sorry. So thank you.

Dr. Castro and Monica for letting me know about that weird start. Good morning to you, to Monica from Chicago. All right. So let's, you know, Harvey has a question here or a comment. So, you know, saying the human factor can make facts subjective. Absolutely. So that is one of the bigger concerns that I think myself and a lot of people are going to have with this kind of voluntary pledge between the U.S. government and the big seven tech companies. So yeah, if you're just joining us, tuning in a little late, just as a reminder, the big news here is there is a voluntary agreement between the kind of seven biggest companies that are providing generative AI. So these chatbots, Google Microsoft, Amazon, OpenAI, Anthropic, and Infliction. So a voluntary pledge or agreement

between all of these companies to essentially play by these rules. and kind of govern themselves and to also submit kind of their newer offerings or new products to a third party before being released to the public and then communicating more and being more transparent with the government and with the public on how these tools work and how they use our data and collect data as well. So a lot to unpack there. So yes, Will Santiago with the comment here. Thank you. Yes, Chicago does rock. Thank you for joining. All right. But let's talk. If you guys have more

questions about this big news between government, the government and the CEOs of these seven companies that's happening today, this meeting's happening today. If you have more comments, still leave them, because I'm going to make sure and go after we're done live here and give you my feedback. But let's talk about what we actually started the show. So Susan has a comment here that's actually a great transition because I said we're going to talk about Apple GPT. There's huge breaking news, but we're still going to talk about it. So Susan with a comment saying, why do I feel Apple GPT is going to be better than chat GPT? All right. Adobe just introduced an entirely new way to create, bringing the power and precision of its creative suite into one conversational experience. Meet

Firefly AI assistant now live in the Adobe Firefly app, the all-in-one creative AI studio. Powered by Adobe's creative agent, Firefly AI, assistant lets you start with your vision, just describe what you want, and shape the outcome as it takes form with the assistant. The assistant orchestrates multi-step workflows, drawing on 60 plus pro-grade tools across Adobe Creative Cloud apps, including Photoshop, Illustrator, Premiere, Lightroom Express, and more to help bring your ideas to life. You can also get started with creative skills, a growing library of pre-built workflows for common creative tasks, like batch editing photos, creating mood boards, portrait retouching, and creating social variations.

Every step the assistant takes is visible so you can refine, redirect, or take over at any time. You stay in the driver's seat as the creative director. Adobe Firefly AI assistant now in public beta. See it today at firefly.adopi.com. There's two billion reasons. I have two billion reasons. That's how many devices Apple has. out right now in the wild.

So now let's transition, guys. I know this was a lot. We're almost going to have two smaller shows in one today. So let's talk about that. Let's talk about Susan's question. Will Apple GPT be better than chat GPT? So first, let's go over the basics of Apple's new GPT, what it is, all of that. So it was first kind of leaked.

So Apple hasn't said anything officially. That's important to know. But this was originally from a Bloomberg, an exclusive Bloomberg report talking that Apple has, apparently has a working version of a GPT model, right? So a large language model that they are training on their own data and that they will eventually release to the public. All right.

So there's a lot of different details here. Let me go through the important stuff, right? So it's using a framework called AJ. Also, it's important to know that it doesn't have an official name yet. Some people are calling it Ajax. Some people are calling it Apple GBT. I'm guessing Apple will call it something else. But for now, those terms are kind of being used interchangeably. The actual framework is called Ajax that it's built on. But yes, a lot of people are calling this Apple GBT. So right now, at least according to initial reports, internal team at Apple is using this. And the expectation is that it will be

released publicly sometime in 2024. The biggest use cases that most people are seeing or assuming, because again, this is all just reporting right now. There's no official details, but they are saying it will be obviously incorporated into Siri. That's huge. And I've talked about this on prior shows about, you know, the original AI smart assistants, You know, your series, your Alexa, is there still a Cortana? Is that still a thing? Here's the thing. If you've been using chat GPT or Bard or anything else over the last couple of months,

and then you go back and like talk to your Alexa or talk to your Siri, it is baffling how bad those are now. That sounds so, it sounds like so judgmental for me to say that. But if you're using GPT, chat GPT or anything on a. ongoing basis, it is almost impossible now for me to get any value personally out of a Siri or an Alexa because we are so used now to these GPT products and offerings being so super smart and being great at responding to our queries, right? So those two billion reasons why this will be more successful. And thank you again for that comment because it is the perfect transition, Susan. So will it be better than chat GPT? I think eventually it will.

It eventually will. Think of Apple's normal business model, right? Let's look at the iPhone. The iPhone was not the first smartphone. It was not the first internet connected smartphone. There was others that were to the party way before the iPhone. So Apple has a tendency, and we've talked about this on the show as well, right? We've said how Apple is lagging behind, you know, and as other companies, you know, Microsoft, I think, has been at the forefront of AI, you know, backing GPT products. They're obviously the largest investor. Microsoft is the largest investor in OpenAI. They're using open AI in their Bing chat. You know, they backed, they backed other, you know, huge AI products. So Microsoft is, I think, leading, which is weird to

say because you always think Google and Apple are leading. Microsoft has been the leader in the AI and GPT space. Google has been kind of shortly following behind. And then you have these other companies, you know, Anthropic Cloud inflection, some of these other kind of smaller, quote unquote, but still very powerful GPT models. So why do I think that Apple GBT will eventually be better? Well, that is Apple's business model. That's what's made them one of the best and most valuable companies in the history of our existence is they aren't first. It might seem like they're first, right? Like if we think about the history of smartphones, it seems like the iPhone was first, but it wasn't.

And even so many features of the iPhone or the iOS or the MacOS that make Apple as successful as it is, very few times are any of those features or functions first? They don't play to be first. They play to be better than the rest. So with OpenAI as an example, if you've listened to the show, you know I'm a huge fan of Chat GTPT. I encourage everyone to replace their Google.com or their default search with OpenAI. I encourage everyone to pay the $20 a month for ChatGPT Plus. No, I don't get paid to say that. It's just foolish. I think that's harsh, but I think it's foolish if you're not paying the $20 a month for chat GPT, because what it's capable of, it is astonishing if you use it correctly, right? So we do a course on that. It's free. There's no upsells. Type in PPP,

email me PPP, D.M. PPP. I'll shoot you the access, right? But I think Apple, GPT or Ajax or whatever it's going to be called will be better eventually. Because, Because once they get it right by not being first to the party, if they integrate it seamlessly into Siri, if you can also still, you know, quote unquote chat with it like you would an I message, you know, so a text message, I do think it has the capabilities or the, or at least the possibility to be substantially better. Because one of the things I've even talked about with chat GPT, as great as it is as great as it is. The user experience and the interface is a little clunky, right? It can be hard to find

your previous chats, you know, only being able to have three plugins. You know, it's even a little hard. They did make some improvements in the plugin store. The performance is a little better, but it's just a little clunky. It doesn't feel intuitive, right? Where anything Apple, if you've used an Apple product, there's no learning curve, right? You know, like my mom has never needed, really, me to explain how to use an iPhone. My mother-in-law has never really had to have basic explanation on how to use an iPhone. But, you know, if you're teaching your parent or a grandparent out there, mostly anyone can pick up an iPhone or an iPad with, you know, maybe a minute of instruction, but chat GPT is a little harder, right? I think he's even,

Google Bard is much better. It's much more intuitive than chat GPT. So that's a long answer there on if it will compete, but I think it 100% will eventually, depending on how Apple releases this Apple GBT or this Ajax. A couple of other things, important things to keep in mind. This is not Apple's only AI play. If you're sitting back there and you're seeing everything that Microsoft and Google and all these other big companies are doing and you're like, oh, Apple's just been, you know,

sitting there doing nothing. Absolutely not, right? It's been widely reported. They're working on a self-driving AI car, right? You have the Vision Pro headset, which they announced and will be coming out. I think that's also either late 2023 or early 2024. So that's, you know, the headset is their biggest, most public AI play, right? Being able to bring that mixed reality, you know, to a headset with essentially VR and AR capabilities. Also, Apple has already announced their other big AI play, which is in health and fitness. So it's not like Apple has been ignoring kind of this new generative AI wave, but so much of this is driven by monetizing because what we've seen as companies like Microsoft or if you get

on the hardware side, Nvidia, some of these other companies. that have been investing more publicly into generative AI have seen their stock prices soar. You know, I talk about it's crazy now how Nvidia now is probably one of the most important companies for the U.S. economy. You know, they're one of the only companies, and they have such a disproportionate market share, of being able to make these computer chips, these GPUs, which is what, you know, the most important AI, they need this GPU. And invidia is the only one, right? now that can keep up with demand of creating a high-quality GPU, right? If that goes away, so many companies and just the U.S. economy could really start to slip, right? But that's why I think Apple

has continued to work on a GPT product, even in the background, even as they have the AI self-driving car, the Vision Pro headset, you know, all their new AI fitness and health initiatives. But you have to play to the crowd. And what we've seen is, generative AI, AI chatbots are definitely not a fad. It is the way of the future. And so Apple knows they're smart. They knew that they would eventually have to release this project, release their GPT, kind of their chat GPT competitor.

And especially how it's baked into the iPhone is going to be paramount with its success. You know, I've said this all along. I was talking to a small group of young entrepreneurs in Chicago here last night about AI. And I said, you know, we had, you know, a 10 to 15 year grace period to kind of, quote, unquote, learn Web 2, right? You know, social media, you know, all these other things, you know, that we consider, you know, the new internet. We had a nice 10 to 15 year time period where, you know, as an individual in your career, you had time to figure it out as a company. Companies didn't have to rush online in the early 2000s. They had time. But now if you're not generating some sort of business, it's hard to be successful,

right? With the GPT and this AI kind of wave, you don't have that long. You don't have 15 years. You know, Apple knows that. Microsoft knows that. But I think it's important for the listeners of this live stream and podcasts to know that as well. You know, we talked about that yesterday with our guest, Corey, you know, he said very, very smart guy who's been very successful in tech and AI over the last decade. He said companies have 18 months where if companies don't figure it out, there's a great chance they're going to get lapped by competitors or go towards shutting down. All right. So to recap, we went over a lot today. I'm going to get to a couple of these comments and questions before we close out. Big, big news in the U.S. governance.

of AI. So the seven, seven big companies, Google, Microsoft, Meta, Amazon, OpenAI, Anthropic, and Infliction have submitted to this voluntary pledge, which we already went over. They're meeting with the president today, Friday, in Washington. So there will be more details on that. We'll see how it pans out. And then our other big topic, like I said, we almost did two shows in one today, Apple GPT coming out, or Ajax, whatever you want to call it. It's going to be extremely interesting. I think the biggest advantage is they always, even, even, even if they're not first, they're usually the best. They see the mistakes everyone else is making, and they have that huge, two billion, two billion reasons why I think they're going to

really crush it is they have two billion devices out there. So whenever they do release this, GPT, you know, whether it's baked into Siri or whatever it is sometime in 2024, they're going to, it's going to be used by up to two billion people. All right. Thank you for tuning in. I'm going to get to a couple of these of these comments here. And as a reminder, if you are listening to this on the podcast, this is live, right? We go live. So in the show notes, make sure to click those and you can join us next time every, every weekday morning, 7.30 a.m. Central Standard. Will Santiago with a question, the best way to regulate AI is AI. Do you agree with

that? Begrudgingly, yes, Will. I know it's tricky, right? To have AI regulate itself. Here's what I think. I think it's smart this first iteration of these companies voluntarily signing this pledge. Because if we were to rate for regulation from the U.S. Congress, I don't think it would go well. It would take, number one, it would take way too long. and I don't think the first attempt at regulation will be worthwhile. I've talked about this before, but go watch some Congress people who have interviewed,

you know, Mark Zuckerberg as an example. They don't understand the internet. So many of them don't. That's not me being mean or passing judgment. Go watch the questions. They don't understand the internet. So you can't expect Congress as a whole, anytime soon to regulate AI.

You can't. I'm sorry, you can't. So at least right now, I do think this is a good first step towards responsible and ethical AI usage is AI and the AI companies regulating themselves. There's no great, there's no great way to say that.

Question, Rastafah, what foundation is Apple GPT built on? Ajax, again, there's not a lot of information. But yeah, initial reporting is that they're building their own internal model, that they're not, you know, using like OpenAI's, you know, API integration like Microsoft is. All right, Dr. Harvey Caster, I predict this will create a new Apple iPhone that will have a small LLM on the iPhone. Absolutely. And that's, you know, there's already been hackers in the, you know, AI and LLM space that have done this that have, you know, essentially put an entire large language model on a phone. which is crazy to think about.

So if you're not a technical user, think of the history of the internet, which is huge when it comes to the file size on your phone and you don't even need to be connected to the internet, right? So that's kind of and the ability to chat with it. So that's kind of what, you know, Harvey was saying there. Rastafah, when was the last time you Googled something? I cannot remember. I'd say most people are still using Google.

Not everyone's like you and I, Rastafah, where we try to, you know, start those searches first in a smart chat, I think. Yes, Nadia with a great point that you don't have to be first to take a big market share. Yeah, Apple is the prime example of that. And also saying Apple will make GPT use seamless and will give users a good experience. Absolutely. Lisa dropping PPP, I'll definitely get that over to you. Let me see.

All right. We have a bunch of other comments. but I hope this was helpful. We talked about a lot today. Thank you so much for joining us. And we're going to talk about this more Monday. All right, we're going to devote an entire show because all of this news was recent. It was breaking.

We didn't, you know, it was an hour ago. So we didn't have enough time to really digest this. So we're going to talk about this again on Monday. So I hope you can join us there for another episode of Everyday AI. Thanks. Meet Firefly AI Assistant. Now live in Adobe Firefly, the Allman One Creative AI Studio. Just describe what you want to create in your own words and the assistant handles the rest,

orchestrating multi-step workflows across Adobe Creative Cloud apps, including Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome while the assistant accelerates execution. Stand control with the ability to step in and refine at any time. See it today at Firefly. adobe.com. And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going. For a little more AI magic, visit Your EverydayAI.com and sign up to our daily newsletter so you don't get left

behind. Go break some barriers and we'll see you next time.