But now the stuff I really was intrigued by: the tech used by Apple Intelligence.
Apple has launched Private Cloud Compute (PCC), a cloud intelligence system designed for privacy-centric AI processing. PCC offers advanced security architecture and stateless computation to ensure data privacy. It coincides with the introduction of Apple Intelligence features, including integration with OpenAI’s ChatGPT. PCC utilizes a custom-built server node with Apple silicon, Secure Enclave, and Secure Boot, providing a narrow attack surface and leveraging code signing and sandboxing for data center security. PCC requests are routed through an Oblivious HTTP relay to conceal IP addresses. Independent security experts can inspect the code running on Apple silicon servers, and software images are published for transparency.
How does this work?
From the keynote:
When you make a request, Apple Intelligence analyses whether it can be processed on device. If it needs greater computational capacity, it can draw on Private Cloud Compute, and send only the data that’s relevant to your task to be processed on Apple Silicon servers.
Your data is never stored or made accessible to Apple. It’s used exclusively to fulfill your request.
And just like your iPhone, independent experts can inspect the code that runs on the servers to verify this privacy promise.
In fact, Private Cloud Compute cryptographically ensures your iPhone, iPad, and Mac will refuse to talk to a server unless its software has been publicly logged for inspection.
If you want to go deep here, MIT Technology Review has a solid look at the technology, and The Register talks about the servers themselves – custom servers based on Apple’s silicon.
Why do we care?
And here’s where things get particularly interesting. This architecture is what all the money buys you – a thoughtful solution built around privacy. Of course it’s sensible to wait for a full security analysis… but just the overview shows the difference in approach from Microsoft, OpenAI, and Google.
What’s also unique about all of this is that it’s a feature, not a product. If you’re an Apple user… you get all this. No extra fees for Apple’s version. That’s not reductive of the IT service provider value, as the advice, guidance, training, and consulting work is all still relevant.
AI is a feature, not a product, and how that feature works is IT’s job to understand and make it relevant for customers.