AI-Powered Blogging: Insightful or Useless?

Ancient history

Way back in 2018, I decided to start a blog. I had some good ideas that not many others were talking about. At the time, I was surrounded by people telling me I wasn’t a real architect because I cared more about business outcomes than about technological innovation.

The engineers who worked for me were hesitant at first. They wanted to use the latest and greatest tools, not just the most reasonable ones. When I left that team for another role in the organization, the technical lead who took over loosened the reins. About a year and a half later, they realized they had let things go too far. They had bet on a technology that had already been abandoned, and they were having frequent delivery problems. The developers were happier, though, at least until the issues started piling up. I never told them “I told you so,” but I certainly thought it.

That blog lasted only about six months before I realized I had two problems:

I was a perfectionist and spent way too long on each article.
Writing takes a lot of time. Good writing takes even more.

So I said goodbye to the blog. I left it up for a while, but I stopped adding to it. I still had a lot I wanted to say. I just didn’t have the time to say it.

Wait, Claude can do what?

Fast forward to today. AI is everywhere, and the practical uses are everywhere too. It’s writing our code, creating our content, even composing our music. It’s amazing, and the pace of change is astounding. If you’re not doing something new with AI every day, you’re getting left behind. Big time.

So, as part of my own continuing education, I decided to resurrect my blog, this time with a partner. His name is Claude Code, and he’s pretty incredible. In just a few hours, he recreated my old blog with all of my old content, gave it a brand-new look and feel, and wrote a batch of new articles. They don’t sound exactly like me, but they’re close, and they’re exactly the kinds of things I would write about myself if I had the time.

That’s what this blog is now: equal parts my thoughts and Claude’s idea of the content that appeals to me. And it’s really impressive. As I read through my “new content,” I realized I was genuinely enjoying it. It agrees with my positions on technology, and it links to resources I never would have found on my own, many of them fantastic.

What started as an experiment with a tool has turned into something more: a training tool I can use to make myself better. Yes, I know the positions it’s espousing lean toward my own, so there’s an element of confirmation bias here. But what is a blog for, if not to share our opinions in a convincing way? Claude is just helping me convince you of mine. Thanks, Claude!

So, what now?

I don’t know how long I’ll keep this going, but right now I’m getting real value from reading the articles Claude writes on my behalf. They’re fascinating, and they cite not only my own (real) writing but also the work of many other great thinkers in the space.

So I’m going to keep going. I will… well, Claude and I will… publish something new every month, maybe more often, and we’ll see how it goes. To keep the experiment transparent, every note gets one of two labels:

AI Generated, Human Approved
Human Generated, AI Improved

This one, for the record, is Human Generated, AI Improved.

If you’re reading this (and you’re not me), I hope that however you found it, you’re finding it at least mildly interesting. I certainly am. Let’s see where this goes.