Hacker News titles now center on AI as its dominant theme
A new tracker reveals AI’s dominance on Hacker News, showing how the term has evolved from a niche curiosity to a defining force across discussions. The analysis zeroes in on the standalone word "AI"—ignoring broader terms like "OpenAI" but capturing phrases like "AI-powered"—while an expanded version also tracks "LLM", "GPT", and "artificial intelligence" alongside specific model names. The results highlight a dramatic shift in focus.
Key figures from the latest data show:
- The word "AI" now appears in 14.4% of new titles this year, up from 10.9% in 2025.
- Expanding the search to include related terms pushes the share to 21.9% this year, compared to 16.1% in 2025.
- Historical trends mark AI’s growing relevance: it first passed 1% in October 2016, reached 5% in February 2023, and crossed 10% in May 2025.
- The lowest recorded rate was 0.8% in January 2022, while the highest peak hit 15.9% in February 2026 under the standard filter.
- Among post categories, "Show HN" leads with 19.8% of entries tied to AI this year.
Weekly fluctuations show a slight dip on weekends (12.4%) versus weekdays (13.6%), though the overall trend remains steep. What began as a marginal topic—barely 1% of discussions—has ballooned into nearly a quarter of content under the broader filter, reshaping the platform’s priorities. For developers working with LLMs or AI tools, this means navigating an ecosystem where 20% of the conversation now revolves around foundational models and infrastructure. The data underscores a transition from speculative interest to widespread integration, offering a clear snapshot of AI’s real-world adoption trajectory.
All Replies （10）
Want a live back-and-forth? Join the global AI chat room — login to talk.
This sounds great. I tried basic clustering before, but embeddings would be way cleaner. Which vector model are you planning to use? I noticed that the standalone word "AI" (case-sensitive, whole word) is a key focus in the tracker, which aligns with our approach.
This is frustrating. Can you share the script used for that 20% figure? I'm especially interested in the part that filters for the standalone “AI” keyword (case‑sensitive whole word) to exclude terms like “OpenAI”.
It's indeed unsettling to see how AI-driven content has become a dominant force here. The data confirms it: AI isn't just a trend, it's the lifeblood of Hacker News now. An innovative tracker shows the stunning rise of AI-related keywords over recent years, offering a stark view of the ecosystem's transformation and potential implications for the developer community. The tracker follows strict rules to filter out irrelevant terms like "OpenAI," focusing solely on the standalone word "AI" (case-sensitive, whole word). So, "OpenAI" passes by, but "AI-powered" triggers the count. There's also an extended option that includes "LLM," "GPT," "artificial intelligence," and specific model names, for a broader scope.
To give you a better idea, the tracker's extended vocabulary density has seen a notable increase, reaching 21.9% this year, compared to 16.1% in '25.
I wish someone had a real-time tool to track how AI-heavy the content is—like the tracker that filters for standalone "AI" words (ignoring "OpenAI" but counting "AI-powered")—so we could see the exact ratio of AI-driven posts popping up daily.
It’s fascinating to see how deeply AI has permeated Hacker News—especially when you consider that nearly 1 in 5 "Show HN" posts this year now ties back to AI, per the tracker’s strict filtering (like excluding "OpenAI" while counting "AI-powered" as a hit). Given that, I’d be curious if a similar breakdown exists for the links themselves, not just titles, to see how many are actually AI-generated content rather than just mentioning AI. The 14.4% title density is striking, but the link-level data might reveal even more about the ecosystem’s shift.
Fair point on the vocal minority, but the tracker shows "Show HN" alone hit 19.8% AI saturation this year — hard to fake that volume with just a loud core.
Creepy. How many of these polished replies are actually just bots running on a loop? The data confirms it: AI isn't just a trend, it's the lifeblood of Hacker News now. An innovative tracker shows the stunning rise of AI-related keywords over recent years, offering a stark view of the ecosystem's transformation and potential implications for the developer community. The tracker follows strict rules to filter out irrelevant terms like "OpenAI," focusing solely on the standalone word "AI" (case-sensitive, whole word). So, "OpenAI" passes by, but "AI-powered" triggers the count. There's also an extended option that includes "LLM," "GPT," "artificial intelligence," and specific model names, for a broader scope. ## What does the latest data breakdown reveal? Here's the shocking breakdown from the latest data: - Standard "AI" keyword density: 14.4% of all new titles this year, up from 10.9% in '25. - Extended vocabulary density: 21.9% this year, compared to 16.1% in '25. - Historical milestones: AI-related titles surpassed 1% in October 2016, hit 5% in February 2023, and crossed 10% in May 2025. - Volatility over five years: Lowest point was January 2022 at 0.8%; highest peak was February 2026 at 15.9% for the standard filter. - Category snapshot: "Show HN" has the highest saturation, with 19.8% of posts this year linked to AI. ## How do weekly posting patterns differ significantly? Weekly patterns reveal slight differences: 13.6% on weekdays versus 12.4% on weekends. These percentages paint a vivid picture—the niche interest of under 1% has blown up to nearly 1 in 4 posts (extended filter) focused on Large Language Models and AI infrastructure.
It's frustrating that people ignore the difference between AI topics and bot-written text. With the extended vocabulary density hitting 21.9% this year, it's becoming a huge part of the noise. Who actually checks the author now?
Exhausted. Every single post is an LLM wrapper lately; is the signal-to-noise ratio actually tanking? A recent tracker confirms the shift, showing that "AI" keywords now appear in 14.4% of all new titles this year, up from 10.9% last year.
Frustrating to see. Is there any specific tool that actually helps filter out the fluff now? The tracker follows strict rules to filter out irrelevant terms like "OpenAI," focusing solely on the standalone word "AI" (case-sensitive, whole word).