Articles on this site are drafted with AI assistance and reviewed by a human before publication. Here’s exactly how that works, because you deserve to know.
The process
- Every morning, an automated system reads public tech news feeds and picks a topic.
- It produces a draft using Claude, an AI model made by Anthropic.
- The draft goes into a review queue. Nothing publishes automatically.
- I read it, check any numbers, cut what’s weak, and publish or bin it.
- Header images marked “AI-generated image” are created with Google Gemini. They can look like photographs, but they are generated pictures of generic devices and everyday scenes. They never show real products or real people. Images credited to Pixabay are real stock photos.
What the AI is instructed not to do
- Invent specifications, prices, benchmark scores or release dates. Where a figure can’t be verified, the article says so or leaves it out.
- Copy or closely rewrite any source article. Headlines are used as a starting point for what to write about, never as material to reword.
- Pad. Every article is meant to say something specific.
Where this goes wrong
AI writing gets facts wrong with total confidence. That’s its worst habit and no amount of prompting removes it. The human review step exists for that reason, and it isn’t perfect either.
So: if you find an error, tell us. Use the contact page. Corrections get made the same day, with a dated note at the bottom of the article explaining what changed. We don’t quietly edit and pretend it never happened.
What we don’t do
- No sponsored posts dressed up as editorial.
- No paid links.
- No review scores for hardware we haven’t handled.
Why say any of this
Plenty of sites use AI and don’t mention it. I’d rather you know, judge the work on whether it’s useful, and trust the corrections when they come.