AI saturation on Hacker News has officially hit a massive

The data is finally in, and it confirms what we’ve all been feeling while scrolling through our feeds: the AI hype cycle isn't just a bubble, it's practically the entire ecosystem now. I was looking at a new tool that tracks the density of specific keywords on Hacker News, and the numbers are actually staggering when you look at the trajectory from the last few years.

The logic behind this specific tracker is pretty strict to avoid "noise" from terms like OpenAI. It specifically counts the standalone word "AI" (case-sensitive and word-bounded). So, if a title says "OpenAI," it doesn't trigger the counter, but "AI-powered" does. There is also an extended vocabulary toggle that pulls in terms like "LLM," "GPT," "artificial intelligence," and specific model or vendor names for a much broader view of the landscape.

Here is how the actual density breaks down based on the latest stats:

• Standard "AI" keyword density: 14.4% of all new HN titles this year, up from 10.9% in 2025.
• Extended vocabulary density: 21.9% this year, compared to 16.1% in 2025.
• Historical milestones: The first time AI-related titles broke the 1% threshold was October 2016. We crossed the 5% mark in February 2023, and we finally blew past 10% in May 2025.
• The volatility spectrum: Over the last five years, the absolute floor was January 2022 at a mere 0.8%. On the flip side, we saw a massive peak in February 2026, hitting 15.9% for the standard filter.
• Category breakdown: Interestingly, the "Show HN" category is the most saturated, with 19.8% of posts this year being AI-related.

If you look at the weekly rhythm, it's not a huge difference, but there is a slight lean toward work-week activity: 13.6% on weekdays versus 12.4% on weekends.

When you're trying to perform a deep dive into what's actually trending in the developer community, these percentages are a reality check. We aren't just seeing "more" AI content; we are seeing a fundamental shift in the topic distribution of the most influential tech forum on the web. We've gone from a niche interest (under 1%) to a situation where nearly 1 in 4 posts (using the extended filter) is directly tied to Large Language Models or AI infrastructure.

It’s a useful metric for anyone trying to understand the real-world deployment of these technologies versus just the marketing noise. If you are building an AI workflow or looking for the next big LLM agent breakthrough, you're essentially navigating a sea where 20% of the water is made of pure silicon and transformer architecture.