At #WWDC26, Apple finally delivered much of what it had promised two years
At #WWDC26, Apple finally delivered much of what it had promised two years earlier.
Many people will look at that and conclude Apple was behind on #AI.
I think they learned something important the hard way.
The bottleneck was never the model.
It was the data.
Before an AI assistant can answer useful questions, it needs context.
It needs to know:
- where information lives
- how information is organized
- how systems connect to each other
- what the user has permission to access
A model cannot answer:
"Find the presentation Sarah sent before the Tokyo trip and summarize the hotel recommendations."
unless the underlying infrastructure can identify:
- Sarah
- the trip
- the emails
- the attachments
- the documents
- the relationships between them
That is not an AI problem.
It is a data problem.
Over the last few years Apple invested heavily in indexing, retrieval, app integrations, semantic search, and system-level infrastructure. Only then could AI become genuinely useful.
Most companies are in the same position today.
They are racing to deploy models while their information is scattered across:
- email
- Slack
- SharePoint
- Google Drive
- Salesforce
- HubSpot
- spreadsheets
- employees' heads
The lesson from Apple is not that better models win.
The lesson is that better context wins.
The organizations that benefit most from AI won't necessarily be the ones with access to the smartest models.
They will be the ones that have done the hard work of organizing, connecting, and making sense of their data.
AI is becoming a data architecture problem disguised as a model problem.