Mike MurphyAI Handyman
Subscribe
Back to issues

AI Unplugged · by · sent January 23, 2026 · Issue 042

Claude’s Constitution

[

🗞️ AI News of The Week: Claude Constitution

[

Read Blog Post

](https://www.anthropic.com/constitution?utm_source=www.therundown.ai&utm_medium=newsletter&utm_campaign=anthropic-writes-claude-s-constitution&_bhlid=1e4e2dc64d0f9db71f20ff014e7295c67a69f16a)

Anthropic has published Claude’s Constitutional AI framework…the actual principles and values that guide how Claude behaves and makes decisions.

Why this matters:

  • Transparency breakthrough - First major AI company to fully publish their AI’s ethical guidelines

  • Democratic approach - Constitution combines input from diverse global perspectives, not just Silicon Valley values

  • Real-world impact - These aren’t theoretical principles; they’re what actually shapes every Claude response

  • Trust building - Shows exactly what guardrails exist and why certain requests get declined

👉 Read More About Claude Constitution

🧠 AI Term of The Week: Agentic

Definition: When AI systems can act autonomously to achieve goals, make decisions, and take actions without constant human intervention.

Explain Like I’m 5: The difference between a calculator (you control every step) and a personal assistant (you set the goal, they figure out how to achieve it).

Real examples:

  • Not agentic: ChatGPT answering one question at a time as you ask

  • Agentic: Claude Code automatically debugging your program, searching documentation, and implementing fixes across multiple files

  • Not agentic: AI generating an image from your prompt

  • Agentic: AI agent researching your brand, analyzing competitors, creating content strategy, and scheduling posts

Why it matters now:

  • We’re shifting from “AI tools” to “AI agents” in 2026

  • Agentic systems can handle complex, multi-step workflows independently

  • This is what enables AI to move from helpful assistant to actual productivity tools.

💬 Quote of the Week:

❝

“Either you run the day or the day runs you.”

– Jim Rohn

🧰 AI Tool of The Week: Hume AI

[

Hume

](https://hume.ai?utm_source=aiunplug.io&utm_medium=referral&utm_campaign=issue-42-claude-s-constitution)

What is Hume AI: Realistic & expressive voice AI models powered by emotional intelligence for creators, developers, and enterprises. Create audio books, podcasts, conversational agents and more.

What it does:

  • Analyzes vocal tone, pitch, and rhythm to detect emotional states

  • Generates AI voices that match and respond appropriately to user emotions

  • Enables more natural, emotionally-aware conversations with AI

Why it’s interesting:

  • Goes beyond words to understand how people say things

  • Creates AI interactions that feel genuinely responsive and human

  • Opening new possibilities for mental health support, customer service, and entertainment

**Real-world example:*****FYI:****This is the tool used by **Inception Point AI *to create all their AI-generated podcasts (that many in the podcast industry are up-in-arms about).

Explore Hume AI

🛠 What I Made: Claude Code: How To Add Trusted Directories

Description: In this tutorial, you will learn how to add Trusted Directories (Folders) to prevent Claude Code from asking ‘Do You Trust Files?’ when launching from a trusted directory.

🤖** What You’ll Learn**

  • Locate Claude Code Settings.json file

  • Add Folder as Trusted Directory

  • Save & Exit Nano Text Editor

  • Launch Claude Code from Trusted Directory 📺 Watch Tutorial on YouTube