Claude Code + iMessage is Finally Here.

In This Issue

• How to text your Claude Code session from your iPhone and get it working on real tasks
• A non-coder who spent 15+ years dreaming of building his product finally started -- in just 3 days
• Every top AI model just scored under 1% on the new intelligence test -- and what that means for you

New Video:

Claude Code + iMessage is Finally Here.

Set up Claude Code's iMessage channel and you can text it tasks from your phone -- it runs them on your machine as if you were sitting right there. Nate covers the full setup (it takes minutes), how to approve permissions remotely from your phone, and when to use this versus Dispatch or Remote Control.

The mistake most people make: Assuming a feature like this requires some complex server setup or tunnel to configure. That assumption keeps you from using tools that are already available and could make you meaningfully more productive today.

How this helps you: Once this is running, you can delegate tasks to Claude Code from anywhere -- no laptop in front of you required. That's a different way of working: one where your AI agent runs jobs while you're out living your life.

Winning Is All We Do

A 15-year dream, built in 3 days

What happened: Ian Darley spent 15+ years trying to build an interactive personal development platform -- one that adapts to each user, keeps them on track, and provides support when they need it. He wanted something more than a course. He thought it would stay science fiction. Three days after discovering Claude Code through the AIS community, he's building it. He's already picking up pace and confidence.

The opportunity for you: If you've had a product idea stuck in your head because you assumed you needed to know how to code, that barrier is gone. Claude Code combined with the right community is letting people build things that simply weren't accessible to them before.

This is one of many success stories from AI Automation Society Plus. If you're not a member, you're missing out.

In the News

Every Top AI Model Just Scored Under 1% on the New Intelligence Test

The news: ARC-AGI-3 launched on March 25 -- a new benchmark that replaces static puzzles with interactive, video-game-like environments where AI agents must discover rules and solve problems with zero instructions. Every frontier model failed. GPT-5, Claude Opus 4.6, and Gemini 3.1 Pro all scored under 1%. Humans scored 100%.

How it affects automation: Current AI is exceptional at executing tasks it's been trained on -- and terrible at learning new ones on the fly. That gap is exactly why automation builders who understand how to structure workflows and guide AI agents will be irreplaceable for a long time. The tool is powerful. The person who knows how to use it still matters.