Blog
Notes on writing, editing, and working with AI without losing your own voice.
How to Use AI for Meeting Notes That Are Actually Useful
AI can turn a transcript into a summary in seconds. The problem is most of those summaries read like a recap of who talked, not a record of what got decided.
Writing Ecommerce Product Descriptions With AI Without Every Listing Sounding the Same
AI helps at catalog scale, but the failure mode is specific: every product in the store starts reading like the last one. Feature lists, platform differences, and the workflow that actually converts.
How to Use AI for Emails Without Sounding Like a Robot Wrote Them
Emails have their own genre rules, different from blog posts or essays. Subject lines, openers, and sign-offs that give away an AI draft, and how to fix each one.
Using AI for Cover Letters and Resumes Without Sounding Like Everyone Else
A recruiter reads a stack of these a day and an ATS scans every one before a human sees it. Here's what actually matters when AI is part of the job application process.
How to Spot AI Writing as a Reader: A Field Guide to the Tells
Not the writer's guide to fixing your own AI draft, the reader's guide to recognizing one someone else handed you. Structural and content-level patterns beyond the usual phrase lists.
How to Get ChatGPT or Claude to Consistently Write Like You
Sounding less robotic and sounding like you specifically are different problems. Here's how to actually capture your own voice and get a model to hold it across sessions.
Which AI Model Writes Best in 2026? It Depends on the Job
A look at how Claude, GPT, and Gemini compare for writing in 2026, based on multiple independent tests, and why the honest answer is not one model, it's the right model for the task.
How Does Claude Watermark AI-Generated Text? What Actually Changed
Anthropic started embedding invisible watermarks into Claude-generated text in August 2026. Here's the actual mechanism, straight from Anthropic's own documentation, what it can and can't prove, and what survives editing.
How to Fact-Check AI-Generated Content Before You Publish It
AI models write confidently whether they're right or not. A practical, repeatable process for catching the mistakes before your readers do.
Does AI-Generated Content Hurt Your SEO? What Google Actually Says
Google's own documentation on AI content and search ranking, what it actually says, what "scaled content abuse" really means, and how to use AI writing without risking your rankings.
How to Prompt AI So It Doesn't Sound Like AI
Most people fix robotic AI writing entirely after the fact. A lot of that work is avoidable upstream, with a better prompt in the first place. Here's exactly what to change.
How to Keep AI Writing On-Brand: Building a Style Guide That Actually Works
Adjectives like "friendly" and "professional" don't tell a model anything useful. Here's how to build an AI style guide that actually keeps voice consistent across a team.
How to Make AI Writing Sound More Human (Without Losing Your Voice)
Why AI-generated text has such a recognizable "tell," and the specific, practical edits that fix it, from sentence rhythm to the phrases that give it away every time.