---
title: "Siri Grows Up"
issue: 76
description: "Apple just gave Siri the overhaul everyone's been waiting for, bringing more capable, more personal assistant features across all of its apps."
date: 2026-09-18
canonical: "https://mikemurphy.ai/ai-unplugged/issues/076/"
agent_json: "https://mikemurphy.ai/ai-unplugged/issues/076.json"
---

## 🌲 Agent Evergreen Build Log: Traveling

Traveling all week in Franklin, TN, so no build updates this time around. Evergreen's still moving. Back at it next week with real progress to share.

- **Waitlist Signups:** 3
- **Read the full manifesto:** [mikemurphy.ai/articles/agent-evergreen-manifesto](https://mikemurphy.ai/articles/agent-evergreen-manifesto/)
- **The Founding waitlist:** [mikemurphy.ai/evergreen](https://mikemurphy.ai/evergreen/)

---

## 🗞️ AI News of The Week: Apple Relaunches Siri With Real AI

Apple just gave Siri the overhaul everyone's been waiting for, bringing more capable, more personal assistant features across all of its apps.

**Why this matters:**
- Apple's finally bringing a serious AI assistant
- Personal by design
- Unlike standalone AI apps, Siri AI ships pre-installed on every iPhone

👉 [Read More About Siri](https://www.apple.com/newsroom/2026/09/siri-ai-a-profoundly-more-capable-and-personal-assistant-is-here)

## 🧠 AI Term of The Week: Grounding

**Definition:**
Linking an AI's answers to real, verifiable source documents to make sure it tells the truth.

**Explain Like I'm 5:**
It's like making a student show their work and cite the textbook page instead of just guessing at the answer.

**Why it matters now:**
- Hallucinations are still the biggest trust problem in AI, and grounding is the main fix
- More tools now cite their sources by default, not as an afterthought
- For creators and businesses, grounded AI means fewer embarrassing made-up facts in client-facing content

## 💬 Quote of the Week

> "Do not take life too seriously. You will never get out of it alive."
> — Elbert Hubbard

## 🧰 AI Tool of The Week: Jev by TypeSafe AI

Jev is TypeSafe AI's first "System One Model," built to make fast, reliable decisions inside software and automated workflows instead of just chatting with a human.

**Why it's interesting:**
- It's a different shape of AI model, built for machines calling machines instead of humans typing prompts
- Points to where a lot of AI infrastructure is heading
- Useful for builders who want AI judgment inside a pipeline without the overhead of a full chatbot

**Real-world example:**
A workflow tool could call Jev to decide whether an incoming support ticket needs a human, instead of routing every ticket through a full LLM conversation.

👉 [Check Out Jev & TypeSafe AI](https://typesafe.ai/blog/introducing-system-one-models-and-jev)

## 🛠 AI Tutorial Of The Week: How To Push an Astro Site to GitHub & Deploy on Cloudflare Workers

**Description:**
In this tutorial, you'll learn how to push an Astro site from your local computer to GitHub and deploy it automatically to Cloudflare Workers.

**🤖 What You'll Learn**
- Wrangler: installing the Cloudflare CLI
- wrangler.jsonc: configuring your Cloudflare Worker
- GitHub: pushing a local Astro project to a new repo
- Cloudflare Workers & Pages: importing your GitHub repo and setting build settings
- CI/CD: auto-deploying on every GitHub push
- Custom domains: connecting your domain to Cloudflare

📺 [Watch Tutorial on YouTube](https://youtu.be/gBd3XsYidpk)

---

**Mike Murphy — The AI Handyman 🧰**
Helping creators & small businesses turn their content & documents into AI-powered tools.