Headlines are loud, decisions are quiet. We walk through the breakthroughs that matter to leaders, agent teams, bigger context windows, and permissionless tools, while Mark brings a CTO’s lens to “knowledge work is cooked” and Rowan shows how to scale marketing without losing voice, trust, or brand integrity.

We cover:

Agent Teams, Bigger Context Windows & Permissionless Tools
Separating Hype from the Shift Already Here
Scaling Marketing Without Losing Voice, Trust, or Brand
OpenClaw at 100,000 Stars: Lessons from Over‑Permissions, Rushed Installs & Real Incidents
Ship Agents Safely: Least Privilege, MFA, Audit Trails & a Living AI Policy
Search Has Changed: Optimise for Citation Inside AI Summaries, Not Just Blue Links
Models That Fit the Work: Frontier vs Open Source, Performance, Cost, Sovereignty, Compliance
The Portfolio Approach: Enterprise‑Grade APIs + Local Models for Regulated Data
Human Judgement First: Define, Constrain, Measure—Then Edit Hard
Your 30‑Day Sprint: Refresh AI Policy, Enable Secure Copilot to Reduce Shadow AI, Rework One Outcome End‑to‑End

If this helped clarify your next move, follow, share with a colleague wrestling with AI adoption, and leave a quick review so others can find us.

Chapters

0:00: Setting The Stakes: AI’s Speed

1:17: Three Big Themes For Leaders

2:08: Marketing’s New Search Reality

3:04: What “Knowledge Work Is Cooked” Means

5:23: Human Judgment As Differentiator

10:06: The OpenClaw Explosion

14:19: Safe Adoption And New Use Cases

16:55: Policies To Curb Shadow AI

19:18: Model Upgrades And Agent Teams

22:28: Open Source, Sovereignty, And Trust