Where's Apple in the AI Race?
Decoding Apple's True AI Strategy
Apple Isn't Missing the AI Race – It's Defining Its Own Lane
The global race for AI is on. All major tech companies, including Google, Meta, Amazon, and Microsoft, are placing their bets on AI. However, one notable name is somewhat missing from the mix: Apple.
In this article, we will explore Apple's AI strategy and how it is positioning the company in a post-AI world.
What does it take to be a successful AI company?
Before getting into Apple's AI strategy, let’s understand what it takes for any organisation to become an AI powerhouse and how does Apple fare across the prerequisites. In order to achieve AI dominance, any organisation needs four things:
Infrastructure - In order to train the AI models, specialised infrastructure is required. If you look at the current set of players, apart from Google, every other player is buying this infrastructure from a third party (mainly Nvidia). Unless Google makes a disruptive discovery or Nvidia gets into some exclusive partnership Apple seems relatively safe
Apple Position - NeutralModel - In order to create a strong AI offering model is the backbone. Right now Apple doesn’t seem to have a strong foundational model, which can challenge the pole position of the leaders.
Apple Position - WeakData - The goodness of any AI could be as good as the data it has been exposed to. The publicly available data provides no differentiation to Apple, but the rich user’s personal data at its disposal can provide Apple an edge as compared to its peers. Apple Position - Strong
Distribution - Even the best AI model in the world is of no use unless it has strong distribution channels to reach out to end users and eventually generate revenue. Apple with its strong sales of iphone & macbook has one of the strongest distribution channels. Also, another point to be noted is that this channel helps Apple to tap in some of the wealthiest users in the world.
Apple Position - Strong
What kind of AI does Apple really need to crack?
Apple’s biggest distribution channel is mobile phones, and this channel doesn’t require them to be an AI giant. While using a phone, a user might be interested in knowing the summary of some emails, get reminded of the birthday of their friends and create a customised message, do a photo search etc. Though the number of tasks could be of multiple kinds, the chances are very less that a user will use their phone to write a piece of code or do detailed research on a topic in medical science.
Hence, Apple needs to ensure that they can handle the user queries, which are varied in number but relatively simpler in the current scheme of AI landscape
Hence, rather than focusing on building the world’s most powerful frontier AI model, Apple will really focus on two things: how to create the best security infrastructure that allows a customer’s personal data to be used by an AI model to provide a hyper-personalised AI experience while protecting the personal data. Secondly, provide the best in class hardware to experience the AI use cases via the app / voice interface.
Hence the priority for Apple to win the AI race is as follows -
Use customers’ data to provide a hyper-personalised AI experience
Create the best user interface to experience AI on handheld devices
Build a large language model
Apple & the internet funnel -
Apple has strong access to the consumer markets via the sales of iPhones and MacBooks. This scale of distribution is not small. Apple sells more than 230mn phones in a year and 20 mn+ mac in a year. Considering the lifespan of an iPhone to be roughly 3 years, Apple is adding more than 160 mn new customers to its ecosystem every year.
Furthermore, it is also worthwhile to look at the strength of this distribution channel. Broadly, the internet funnel can be broken into three tiers -
Device & OS: Devices & OS define the top tier of the funnel. Without a device or an OS a user can’t even connect to the internet e.g. - Windows, iPhone
Access Gateway: This refers to a layer that sits in the middle. With an active internet connection, a device and access to a gateway such as a browser or internet applications, a user can access different online destinations e.g. - Google Chrome, Apple Play Store
Service & Content: The third and last layer refers to the final destination where the required services and content sits. Taking the help of the first two layers, a user can access the content or service of her/his choice. E.g. - YouTube
In order to use AI applications, access to a device is the primary step. If we look at the broad internet funnel, Apple sits at the top of the funnel. A strong distribution network and the pole position in the internet funnel give them an advantage, but not a very important one. This advantage is limited by the kind of use cases a user will try on their handheld devices.
Apple has one of the strongest distribution channel but it will not help it enough to be an AI giant
Apple’s biggest distribution channel is mobile phones. While using a phone, a user might be interested in knowing the summary of some emails, get reminded of the birthday of their friends and create a customised message, do a photo search etc. Though the number of tasks could be of multiple kinds, the chances are very less that a user will use their phone to write a piece of code or do detailed research on a topic in medical science.
Hence, Apple needs to ensure that they are able to handle the user queries, which are varied in number but relatively simpler in the current scheme of AI landscape
Hence, rather than focusing on building the world’s most powerful frontier AI model, Apple will really focus on two things: how to create the best security infrastructure that allows a customer’s personal data to be used by an AI model to provide a hyper-personalised AI experience. Secondly, provide the best-in-class hardware to experience the AI use cases via the app/voice interface.
Hence, the priority for Apple to win the AI race is as follows -
Use customers’ data to provide a hyper-personalised AI experience
Create the best user interface to experience AI on handheld devices
Build a frontier large language model
The priority might be a work in progress behind the scenes as it requires finer detailing, and while it is a deal breaker, it can’t be perceived by the customers easily at first hand.
Introduction of liquid glass UI could be seen as the second step where Apple is trying to set itself in a different league by providing a differentiated experience of using devices
The third priority, which is to work on a frontier large language model, might never get worked on
Why partnering with Open AI / Perplexity is a good strategy
Given that realistically, Apple has to support basic AI use cases, in order to make its devices AI potent, it is a very wise strategy to collaborate with some of the leading AI companies to get the intelligence while putting Apple’s entire focus on crafting the best UIs for using AI. At the same time, focus on creating the industry's best protocols for using the customer data to generate a hyper-personalised AI experience
Also, the partner has to be someone who doesn’t have a hardware/devices business for obvious reasons.
What can we expect from Apple next?
While a lot of other companies are trying to perfect the frontier models, we can expect Apple to further strengthen its distribution channels.
This strengthening could be either through launching lower-priced iPhones which support Apple intelligence (e.g. - the launch of iphone 16e) or even getting into more and more home automation devices.
For More Such Detailed Case Studies
Be the Top 1% PM | YouTube | PM Cohort & Resources | Crack PM Interview | AI/Tech for Everyone
About Author
Shailesh Sharma! I help PMs and business leaders excel in Product, Strategy, and AI using First Principles Thinking. For more, check out my Live cohort course, PM Interview Mastery Course, Cracking Strategy, and other Resources
 | A guest post by
|