Essays on AI Automation Strategy, ROI, and Implementation

Get AI Insights in Your Inbox

We send (hopefully) timeless essays, subscriber-only Hot Takes, and quick tips about how to use common AI tools better.

You can unsubscribe at any time.

If It’s AI vs Humans for Design, My Money Is on Humans

A side-by-side comparison of Claude’s AI-generated designs with human work—and why strong designers still create more distinctive, trustworthy brands.

Most Companies Get AI Backwards

A practical three-step AI automation process: define the workflow, automate it end to end, and begin with the employees best equipped to use it.

When AI Costs More Than It Saves

Token-based AI billing can create unpredictable enterprise costs. Here’s why bounded automations are easier to price than open-ended AI development.

Uber Blew Its AI Budget

Why Claude Code and other AI coding tools create unpredictable enterprise costs—and why task automation is a better fit for most businesses.

OpenClaw & Recognizing the Next AI Hype Cycle

OpenClaw’s rapid rise and decline offers a useful test for AI hype: does the technology create enough value to change how people actually work?

Claude Mythos and Anthropic’s Pronouncements

Anthropic says Claude Mythos found thousands of critical vulnerabilities. Is Project Glasswing responsible AI security—or another dramatic marketing campaign?

AGI Has Been Achieved

By one widely used definition, AGI already exists. The harder challenge is organizing companies, workflows, and human oversight to use it effectively.

Efficiency Improvers Are the AI Buyers Who Win

The organizations generating the strongest AI ROI begin with operational bottlenecks—not whichever tool has the most impressive demonstration. Learn how Efficiency Improvers define complete workflows, prioritize measurable impact and select technology only