Despite an initial decline, shares of Apple (AAPL) have risen as investors digest the tech giant’s AI announcements. New Street Research technology infrastructure analyst Antoine Chkaiban joins Market Domination to discuss
Chkaiban noted that while Apple’s Worldwide Developers Conference (WWDC) had “boring” moments on Monday, “it was exciting because we thought it really captured the promise of generative AI at the core of the operating system.” However, he doubts that the technology alone can kick-start a product cycle, explaining: “We’re not really convinced that on-device AI will enable breakthrough user experiences because the hardware simply isn’t breakthrough… since GPT- 4 is readily available on most iPhones via the ChatGPT app anyway, the question is: do users really need to upgrade to use that technology? The answer so far is not really.
Regardless, he believes the stock is “fairly valued,” especially as services are at an all-time high.
For more expert insight and the latest market action, click here to watch this full episode of Market Domination.
This post was written by Melanie Riehl
Video transcription
Um, let’s talk more about this move that we’ve seen as analysts have cheered Apple’s AI announcement.
It’s a bit mixed.
It took a while for them to get excited.
Shares initially fell after Monday’s keynote before rising to record highs.
Joining us the next day is Antoine Kon, Antoine, a technology infrastructure analyst from New Street Research.
Good to see you again.
Um. So what do you think of this kind of delayed reaction we got in Apple stock to all the announcements?
Yes.
Hi Julie.
Thank you so much for having me.
Um So yeah, I’d say Monday’s keynote was boring and inspiring at the same time because on the one hand they presented exactly what the iPhone experience should be when it comes to using Siri uh in general, coming up with of a plan uh combining like an Uber trip and a boarding pass, an address in the calendar item, and so on and so forth.
All of these experiences are frustrating at best today and they all suddenly seemed very tight and very pristine in Monday’s keynote, but that’s kind of deja vu, right?
And on the other hand, it was exciting because we thought it really captured the promise of generative AI at the core of the operating system. It really has the power to redefine the way we interact with technology uh content creation tools with uh like instructions coming in as natural language context is now taken into account uh also in the way the US interacts with the user interacts, er creativity with the US, generating content and performing tasks.
Um Presumably with minimal instructions.
So yeah, that’s why our thoughts were mixed when we watched the Keynotes on Monday.
Uh and I think the stock reaction originally reflected that as well.
Um, I think it also shows that Apple is working really hard on this.
And I think the fact that it could drive a product cycle is still an open question.
It may be what is driving the shares now, perhaps this also gives confidence.
Um The market is gaining confidence in the fact that it can drive a product cycle.
Maybe that’s what we’re seeing in the stock now.
But do you think Antoine, do you think it’s too early to say whether it will drive a product cycle?
Yeah, because I mean, I really don’t really have a good answer for you because on the one hand you have Apple intelligence that will only be compatible with, um, you know, the products that are based on the uh A 17 pro and M one chips and above.
So that’s maybe about 10% of Apple’s active install base.
So if features are well received, it may encourage replacement as only 10% are compatible.
But on the other hand, it’s a great gift, because for now we’re not really convinced that on-device AI will enable breakthrough user experiences.
Uh Because the hardware is simply not groundbreaking.
Uh. You don’t really see an inflection point, you know, in the calculation that I calculate that’s in the phones.
Uh It’s growing, of course, but there’s no turning point.
And since GP T four is easily available on most iPhones via chat one way or another, via the Chat GP T app, the question is: do users really need to upgrade to use that technology?
Uh The answer so far is not really.
So we’ll have to see if the integration with the operating system is truly a game-changing experience.
So Antoine put all that together.
What do you think this means for further increase or not?
For Apple shares?
Yes.
So if you take a step back and look at Apple stock, I think the stock is fairly valued.
You know, the reality is that growth has peaked on most phones.
Uh iphone peaked in December 2021 in September 2022 with uh you know the stock gains over 2029 2019 to 2022 driven by Huawei’s payout in China.
Um. And the transition to proprietary silicon is now largely behind the macrons.
Then you have iPad which peaked in December 2020 in installed base growth.
Now the cycles are getting slower and longer.
So if you look at services they’re not that high at all and that’s driven by the App Store which is driven by licensing or the deal that they have, with Google and um that collection of companies is probably going to continue to grow double digits in total, but here too we see limited room for acceleration.
So the bottom line is that Apple may have a mid to high single digit growth outlook.
Um, and it’s trading at a very significant premium to the names S and P uh to SNP, with a similar growth profile.
So yes, we still have a bit of and and the product cycle driven by AI, as I said before.
So put that all together and um, and that’s why we’re neutral on the name.
Do you notice Antoine, I would like to point out that Apple’s keynote was music to the man’s ears.
How come Antoine?
Yes.
Well, if you, and actually, if you look at the implications of the event as a second-order derivative for the broader technology infrastructure supply chain, then there’s poor, but I mean, Apple is going to have to pay as it is I open AI will have to pay to get uh uh GP T of O into the US.
And we estimate that Apple should cost about $10 per device per year.
So in the short term, if it’s only 10% of the 2 billion installed devices, that’s 2 billion per year, totaling $2 billion per year.
But in the long run, if all installed users adopt these types of models at scale, it might cost Apple $20 billion on its entire installed base of 2 billion devices.
And so, if they want, if they want to have access to enough computing power to support such a large installed base, they, themselves at Microsoft, are going to have to buy a lot of GP US.
So they might have to spend tens of billions of dollars on GP US.
So first of all, before we talk about poor, this really reinforces our, our, our positive bias at NVIDIA.
And then for arm, well, it pushes up the arm-based silicon content in the phones because you have the model part of the technology, the technology is driven by an inference on the device.
Um. And so that’s obviously positive for arm and then uh as it’s positive for NVDR uh and as you know the gray CPU that NVD I’m going to sell with uh with Blackwell but they’re already studying with Hopper is also based on uh on the arm instructions in architecture.
So that is also positive for Arm Antoine.
Super interesting shot, I appreciate your time.
Thank you very much for joining us today.
Thank you.