Why I Still Write Technical Blogs Even in the Age of AI

Lately, I’ve been thinking about something: With AI, almost any technical question can get a decent answer in seconds. So why would I still spend time writing a blog? Who am I writing it for, and what’s the point? I didn’t convince myself immediately. Instead, I decided to talk it over with AI. Here’s a summary of that conversation—the questions are mine, the answers are AI’s analysis, plus my own reflections. --- The Value of Blogs Has Changed, But It Hasn’t Disappeared Me: Knowledge is so easy to get. Do technical blogs still matter? AI’s perspective: The core value of blogs has never been just “storing knowledge.” It’s about capturing experience. AI can deliver facts quickly, but it cannot convey the pitfalls you’ve faced, the trade-offs you made, or why you chose one path over another. My reflection: The articles I most often save aren’t full documentation—they are posts like “Here’s how I solved XX problem.” Stories with context and personal judgment are what really have reusable value. --- Question-and-Answer Logs Are Writing Material Me: I’ve been trying to record my conversations with AI. Can this count as a blog? AI’s perspective: Yes. Q&A has a natural structure—each question addresses a focus point, each answer shows a thought process. It honestly shows how thinking happens: what you care about, how you iterate on questions, what you accept, and what you discard. My reflection: This post is an example. I didn’t start with a conclusion; I discovered what mattered to me by asking along the way. Hiding this process would lose half the informational value. --- If Multiple AI Answers Agree, Can I Trust Them? Me: If I ask several AI models the same question and the answers are similar, is it more reliable? AI’s perspective: Not necessarily. Different models share overlapping training data, so “agreement” can just be a shared bias. Cross-validation helps for questions with clear answers, but for edge cases or judgment-based questions, consensus doesn’t guarantee correctness. Verification against documentation, experiments, or real results is still necessary. My reflection: I treat AI answers as drafts, not final conclusions. They help me get started faster, but the verification step is crucial. Writing that process down is a reminder to readers: don’t copy blindly—think it through first. --- Knowledge Is Cheap, But Judgment Is Scarce Me: AI provides so much knowledge. Is there still value in what I summarize? AI’s perspective: Yes, and increasingly so. Knowledge acquisition is nearly free, but the ability to evaluate and organize it becomes rarer. Your blog records not just answers, but your question awareness, filtering logic, and practical experience—things AI cannot generate for you. My reflection: It’s not the knowledge itself that matters most; it’s how I use it. Same knowledge applied differently yields completely different conclusions. That’s the part only I can write. --- How to Write a Blog Without Wasting Time Me: If I organize Q&A into a blog, is there a recommended structure? AI’s perspective: A good Q&A blog shows the complete thought process, not just conclusions. A suggested order: 1. Background and Motivation: Why is this question worth asking? 2. Questions and Iterations: How were questions raised and refined? 3. Filtering and Verification: Which answers were accepted or discarded, and why? 4. Extracting Methods: Are there reusable approaches or frameworks? 5. Supporting Materials: Code snippets, screenshots, or diagrams to aid understanding. My reflection: In practice, just clearly showing “what I encountered” and “how I thought about it” is enough for readers to follow. You don’t need to make every post a full methodology paper. --- At the End, Did I Convince Myself? Mostly, yes. I used to worry that blogging would become just “repackaging AI answers” with no real contribution. But now I see that the true value lies in: what you experienced, how you judged, and which pitfalls you encountered. AI doesn’t have that, and search engines can’t provide it—it only exists in your own head. Writing it down ensures it survives beyond memory. --- This post is organized from a conversation with AI. Questions are mine; answers are selected and refined with my own insights.