AI Search news and analysis on Web Pulse: 728 curated articles tracking the latest AI Search developments, tools, and research, updated continuously from vetted sources.
Hybrid search fails when teams merge dense and lexical results by min-max normalisation, because cosine distance is bounded while Postgres ts_rank_cd and BM25 scores are unbounded and shift as the corpus grows, according…
Google updated its helpful content search documentation to warn publishers against using deceptive authorship information, stating that fabricating creator profiles with AI-generated headshots, made-up names, or false cr…
Advocentral analyzed 8,601 US law firm homepages from the September 2026 HTTP Archive mobile crawl and found that three crawl-derived figures were wrong until verified by hand. The robots_txt metric's 6.6% AI-crawler Dis…
SerpApi released an open-source plugin for OpenClaw, the self-hosted open-source personal AI assistant, connecting OpenClaw's built-in web_search to SerpApi and adding 30 specialized tools for news, maps, shopping, fligh…
SEO consultant Sara Taher published Part 2 of her newsletter series on building a search engine, covering the main content extraction step of indexation. Taher used Claude Code to extract text from only the 541 pages tha…
Constructor announced Agentic Checkout on October 5, a Stripe-powered feature that lets consumers complete purchases directly inside an onsite AI shopping agent, live at launch retailers including Sephora, Under Armour, …
A technical explainer on knowledge cutoffs and parametric memory states that Allen-Zhu and Li (2024) measured a language model's knowledge capacity at about 2 bits per parameter when each fact appears roughly 1,000 times…
A developer benchmarked Exa and Tavily's agent-oriented search APIs on a fixed suite of 300 company-news questions, finding Exa fast reached 99.3% answer accuracy versus 87.7% for Tavily basic and 93.0% for Tavily advanc…
A developer's benchmark of three local embedding models on a 16 GB M4 Mac found Qwen3-Embedding-4B (Q4 GGUF) reached 91% natural and 71% paraphrased top-5 retrieval accuracy on 56 job-posting questions against a 400-post…
A six-point self-audit published by an unnamed technical blog gives developers pass/fail curl checks for the files AI agents from Claude, ChatGPT and Gemini read to discover and call services, covering llms.txt, the A2A …
A developer tuning web search in Open WebUI switched from a self-hosted SearXNG port to Brave Search's API, citing Brave's price of 0.5 cents per search versus Kagi's 1.2 cents and Brave's 1000-query-per-month free tier.…
An analysis of 173,020 URLs by Surfer SEO found that pages ranking for related fan-out queries were 161% more likely to be cited in Google AI Overviews than pages ranking only for a main query. The company's guide recomm…
A developer outlined a Node.js contract for combining lexical and semantic search signals when enriching game catalog documentation, keeping embedding-provider response shapes out of the domain model. The approach stores…
DaVinci Commerce launched a product-memory layer for AI shopping agents on October 1st, combining Product Context Memory, a Brand Agent and an Agentic Discoverability Engine, the company announced on October 5th via PR N…
Google Ads will automatically identify and extract high-quality promotions from advertisers' websites and apply them directly to Search and Performance Max campaigns with linked location assets starting October 12, 2026,…
SerpApi released its weekly changelog for September 28 to October 4, 2026, adding a top_news array to its Google News Results API and rebuilding reference parsing for Google AI Overview and AI Mode results. The reference…
A developer built an official WordPress connector for Cognee (PR #221) during a hackathon, letting the AI memory engine ingest blog content as an interconnected knowledge graph rather than isolated vector snippets. The c…
The European Commission designated ChatGPT a Very Large Online Search Engine under the Digital Services Act on Aug 31, 2026, citing its hybrid prompt-and-search nature and 159.1 million average monthly EU users, well abo…
A study of end-to-end search with LLM agents finds that the agents develop a preference for items based on their source, shaping which products, hotels, or papers users receive and which sources get selected. The researc…
A vector search tool alone cannot answer counting or filtering questions over video archives, according to a technical report that found fixed-length clip chunking raised an NFL-game hallucination rate from 68.1% to 74.2…