Skip to content
Apple · OpenAI · SamsungEpisode 011 · 23 June 2024 · 33:20

Apple Has Put AI Into the iPhone, but the Biggest Risk Is Not Siri

Central question

What risk appears when Apple embeds AI into the iPhone more deeply than the user can see or control?

What you take away

Assess the price of deeper AI integration in the iPhone: which actions become easier, which data the system receives, and where the user loses the ability to inspect or reverse a decision.

Main threads

What to watch for

1Compare “ToTheMoon — a podcast about modern technology” with “The point is infrastructure and privacy, not generative emoji”: they provide different criteria for judging the same issue.
2Test the conclusion from “Why people choose Apple. How the AI integration can "fail" because of OpenAI policy” in your own use case—what actually changes in the process and what remains a promise.
3Before choosing a product or approach, record the constraint identified in “Stagnation in the mobile-device market. Is something finally moving?”.
4Define the owner of the outcome and the quality metric for the situation described in “The M4 processor”.
Signals to track afterwards
→Watch for actions by Apple and OpenAI that confirm or challenge the episode’s central claims.
→Compare new launches and policy changes with “Why people choose Apple. How the AI integration can "fail" because of OpenAI policy”: have access, quality, price, or constraints changed?
→Check whether the scenario in “The M4 processor” becomes repeatable practice rather than a one-off demonstration.
Most useful for
EntrepreneursProduct teamsAI usersEveryday usersCompaniesSecurity specialists

Key takeaways

00:00For a long time, Apple looked a company that had missed the beginning of generative AI

The “ToTheMoon — a podcast about modern technology” scene leads to a working conclusion: Apple unveiled its own AI system alongside integrations with outside models — and, joining a year later than everyone, the incumbent effectively "killed" a raft of startups, from Grammarly to consumer image generators.

04:33Why an announcement is not enough: the graveyard of hardware startups

The decision in “The graveyard of hardware startups” depends on one criterion: Rabbit R1 and similar "middleman" devices between the user and generative models have ridden into the sunset: the phone you always carry covers their tasks out of the box, and none of them used the year of head start before Apple joined to actually accumulate users.

05:46Apple's main bet is on infrastructure and privacy

The “The point is infrastructure and privacy, not generative emoji” topic becomes clearer once this point is included: with Private Cloud Compute Apple says outright: "we are the only ones able to analyze all your data — mail, photos — without handing it to anyone, while third-party systems will see it"; that matches the main reason people choose Apple in the first place.

08:53The boundary between value and constraint: will Siri get smarter

The practical meaning of “Will Siri get smarter?” is that the permission button stops protecting after the sixth automatic "Yes", and why call ChatGPT through Siri instead of the app remains an open question; what impresses is different: search inside videos ("show my son throwing a ball a month ago") and a seriously rebuilt system underneath.

11:13But this creates dependence on OpenAI

For the “Why people choose Apple. How the AI integration can "fail" because of” scene, the decisive point is this: data-retention policies, workspace security, and access rules can change outside Apple's control: if Siri effectively hands a line of text to an external system, a polished interface does not remove the questions of what exactly left the device, where it was stored, and who now answers for the result — the integration may fail through lack of trust, not a weak model.

16:43What changes in real work: apple vs. Android? Will Copilot+PC win the

The decision in “Apple vs. Android? Will Copilot+PC win the standoff?” depends on one criterion: the Copilot+PC presentation showed nearly the same AI features earlier; the strength of Apple's update is not image generation (emoji as the "headline feature" strikes the hosts as nonsense) but the combination of many applications with a large infrastructure underneath.

20:48Google and Samsung are moving faster because they are willing to experiment directly inside Android

The practical meaning of “Stagnation in the mobile-device market. Is something finally moving?” is that microsoft is connecting Copilot to a new generation of computers. Apple is moving more slowly, but it is trying to build a single layer between the device, personal data, and several models. That gives the company a chance to avoid choosing one permanent winner and to change suppliers without changing the interface for the user.

24:26How the issue moves from news to product: elon Musk bans iPhones

The decision in “Elon Musk bans iPhones” depends on one criterion: the hosts read Musk's "I will ban Apple phones for my employees" as hype in the spirit of "ban today, allow tomorrow" — especially since Apple is so closed that no detail can be pried out of its staff; the outcry itself only underlines how serious the fight over data has become.

27:50The practical meaning of the issue: why nothing is clear right now

The discussion of “Why nothing is clear right now” yields a practical test: the presentation itself contains only the line that ChatGPT can be summoned from inside Siri — as on any Alexa — while the news already diverges from it (for example, "free ChatGPT for everyone", which the presentation never said); it is still unclear how calling it through Siri beats simply opening the app.

30:17Elon Musk's calls to ban the iPhone only underline how serious the fight over data has become

The decision in “The M4 processor” depends on one criterion: the iPad already carries the M4, a clear step past the M3 — and the real question is no longer the spec numbers but which AI tasks a device with such a processor can run on the spot, without the cloud.

What this episode is about

Apple promises to connect the iPhone with ChatGPT and Gemini while keeping control of personal data. But relying on outside models makes the company dependent on rules set by OpenAI and Google—and that is where the entire idea of Apple Intelligence could break down.

For a long time, Apple looked like a company that had missed the beginning of generative AI. Now the difference between an old and a new iPhone may become noticeable again: the system is supposed to understand context, work with data on the device, and call an external model only when its own capabilities are not enough. That matters far more than the generative emoji that fill a presentation so easily.

Apple's main bet is on infrastructure and privacy. Before sending a request to ChatGPT, Siri is supposed to ask for permission, while some tasks are planned to run locally or in a protected cloud. That approach fits one of the main reasons people choose Apple in the first place: users do not want to investigate which company received their message, photo, or document.

But this creates dependence on OpenAI. Data-retention policies, workspace security, and access rules can change outside Apple's control. If Siri is effectively passing a line of text to an external system, a polished interface does not remove the question: what exactly was sent, where was it stored, and who is now responsible for the result?

The integration may fail not because the model is weak, but because users do not trust it.

Google and Samsung are moving faster because they are willing to experiment directly inside Android. Microsoft is connecting Copilot to a new generation of computers. Apple is moving more slowly, but it is trying to build a single layer between the device, personal data, and several models. That gives the company a chance to avoid choosing one permanent winner and to change suppliers without changing the interface for the user.

Elon Musk's calls to ban the iPhone only underline how serious the fight over data has become. The war between Apple and Android is no longer about the camera or the shape of the phone. The winning ecosystem will be the one that gives people useful AI while convincingly explaining why their personal information did not become the price of convenience.

The war between Apple and Android is no longer about the camera or the shape of the phone. As a result, the winning ecosystem will be the one that gives people useful AI while convincingly explaining why their personal information did not become the price of convenience.

Episode transcript

The episode is in Russian; below is an English reading guide to the transcript (the full EN transcript is a machine translation). Voice matching applied to 80 segments: 41 identified, 4 mixed, 14 probable, and 21 unresolved.

Read transcript on a separate page

Loading…