How Should the Siri App Be Graded?
Summary
John and Chris review the final iOS 27 betas, focusing heavily on Siri AI. They critique its shallow responses and limited app integrations compared to superior free alternatives like ChatGPT’s Luna model. While praising macOS Golden Gate’s performance and Spotlight improvements, they demonstrate how advanced AI agents outperform Siri in complex tasks. The episode concludes with a discussion on using Codex and Fable for sophisticated automation workflows.
Topics discussed
Intro and iOS 27 Beta 5 status
Why this beta season feels boring
Siri AI vs. ChatGPT and Gemini
User behavior and AI adoption trends
Siri AI response quality and genericness
Impact on web traffic and tutorials
App intents and Mac Finder limitations
ChatGPT tool use vs. Apple's approach
Personal context and iMessage search issues
Spotlight indexing vs. Siri AI search
Spotlight improvements in macOS Tahoe
Demo: Siri AI vs. Codex iMessage search
Siri AI is two years behind the curve
Federico's complex Shortcut automation
Federico's multi-Mac Codex setup
Running local AI models on Mac Studio
Device clustering and remote management
Listen ad-free on Castria