Free AI audit
Free AI audit
How assistants pick the sources they quote, what actually changes when one question becomes ten, and what you can do about it on your own site. 20 articles in English.
30 copy-ready ChatGPT Images 2.0 prompts for storyboards, infographics, floor plans, UX mock-ups and branding, plus the five-part template behind them.
Read the article September 2026Claude Fable 5.1 cuts token costs by 25-45% and OpenAI's Astra crosses the Critical cyber threshold. What both mean for SEO, GEO and AEO visibility.
Read the article September 2026Which AI crawlers feed training data and which feed live answers, how to control each one in robots.txt, and why blocking everything by default backfires.
Read the article September 2026Retrieval systems cut your page into chunks and judge each one alone. How chunking works, which formats survive the cut, and what silently kills a section.
Read the article September 2026What Google AI Mode is, how it differs from AI Overviews, how query fan-out selects sources, and what the Search Console generative AI report shows and hides.
Read the article September 2026Grounding anchors an AI answer in retrieved web results instead of memory. How Gemini, ChatGPT and Perplexity do it, and what makes a page citable.
Read the article September 2026llms.txt is a proposed markdown map of your site for AI models. The format, who actually reads it, and whether adding one is worth your afternoon.
Read the article September 2026Since 2021 Google can rank a single section of a page on its own. The same habits that win passage ranking also win RAG retrieval in AI search. Here is how.
Read the article August 2026Does robots.txt block Googlebot, and does a 30-page site need a sitemap? Crawl blocking vs index blocking, five common mistakes and a five-minute test.
Read the article July 20265 must-have SEO skills for Claude Code and 5 GitHub repos that speed up your SEO work - from Google Lighthouse to our own neurise-seo package inside neurise-adseo.
Read the article June 2026Four ways to use Claude Code from your phone: Anthropic's Remote Control, Orca, Tailscale and tmux. How they compare, setup steps and the security trade-offs.
Read the article June 2026Anthropic reported 80.3% for Claude Fable 5 on SWE-Bench Pro vs 58.6% for GPT-5.5. How much of that lead survived GPT-5.6, and how to test both yourself.
Read the article June 2026Grok Imagine Video 1.5 renders sound in the same pass as the picture. We compare xAI and OpenAI list prices and test the 86 per cent cheaper claim.
Read the article June 2026NVIDIA DGX Spark runs local LLMs of up to 200B parameters in 128 GB of memory for $4,699. Where the speed limit sits and who it actually suits.
Read the article June 2026A harness is the layer of tools, instructions and rules around an AI model. Self-harness lets the agent improve that layer itself. How the loop works.
Read the article June 2026Vercel eve framework explained: an agent is a folder of files, with durable sessions, a sandbox and approval gates built in. What it offers, and its limits.
Read the article June 2026Loop engineering moves the lever from writing prompts to designing the loop that writes them. Four conditions, five building blocks and where loops fail.
Read the article June 2026Query fan-out is how an answer engine splits one question into sub-queries. In RAG it shapes which passages get retrieved. How to plan for it.
Read the article June 2026SEO, AEO and GEO compared row by row: where you win, who reads the content, how success is measured and which signal moves it. Plus a four-question test.
Read the article June 2026Models quote sources they treat as trustworthy: communities, sites full of real opinions and pages that answer directly. How to get your brand into that set.
Read the articleThe Polish blog carries far more: over a hundred articles on SEO, GEO and AI search. Browse the Polish blog, or write to seo@neurise.io and we will answer in English.