Apple Intelligence
System-level AI that splits work between on-device and private cloud
WHAT IT IS
Apple Intelligence is a set of system-level AI features Apple introduced in June 2024, building generative abilities into iOS, iPadOS and macOS. Work an on-device model can handle is done locally, and requests beyond that go to “Private Cloud Compute” on dedicated servers. The features span writing tools, notification summaries, image generation and an enhanced Siri.
Why it matters
It proposed an architecture that divides work between on-device models and Private Cloud Compute, placing generative abilities at the level of the operating system rather than inside a standalone app.
Key specs
- Form
- System features across iOS, iPadOS and macOS
- Architecture
- On-device models working with Private Cloud Compute
- Modality
- Text, image, audio in; text, image out
Capabilities
Related concepts
Model Compression
Make a model smaller, faster and cheaper with almost no accuracy loss — but you can usually have only two of the three at once
Inference Optimization & Serving
Training happens once; inference happens a billion times a day — and serving is torn between fast first tokens and high throughput, which usually pull against each other
Safety, Alignment & Prompt Injection
A model optimises the proxy we wrote into the loss, never the thing we actually want — the gap between them is the whole alignment problem
Prompting & Alignment
Making a model helpful, honest and harmless is harder than simply making it bigger
Comparable products
Gemini
2024One chat entry point that gathers search, office apps and a multimodal model
Microsoft Copilot
2023Conversational AI woven into the operating system and office apps
ChatGPT
2022The chat window that put a large language model in everyone’s hands
Siri
2011The early voice assistant that brought spoken control to mainstream phones