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.
HN Sieve launched as a tool that automatically discovers, analyzes and scores every Show HN submission on Hacker News using AI, evaluating each project on a 0-10 scale across Innovation, Technical Quality, Documentation …
An engineer outlined a pattern for bounding vector search latency by giving retrieval a smaller deadline than the caller's request, canceling it via AbortController in Node.js or context.WithTimeout in Go, and falling ba…
A developer built an AI citation tracker that scores how well web-grounded models from Gemini and OpenAI identify and describe a tracked entity, and found that a blended visibility score of 29.2 on 21 October concealed w…
A developer outlined a diagnostic approach for vector search latency that spikes under load while median latency and CPU usage stay flat, recommending teams hold the query and corpus constant and test a small matrix acro…
A developer built and deployed a Python RAG knowledge assistant for operational support documents, hosted as a Streamlit demo with source code on GitHub, that answers questions with a supporting quote and cited document …
Google Cloud made BigQuery ObjectRef generally available, letting teams store pointers to Cloud Storage images, video, audio and documents as a native STRUCT column alongside relational data and pass multimodal, multi-co…
Elmo, an MIT-licensed AI visibility tracking tool at version v0.7.3, is being positioned as a self-hostable alternative to Otterly.AI and Peec AI for marketers monitoring how AI answer engines mention their brands. Elmo …
Across 810 AI answers analyzed, vendors received only 2% of citations when buyers asked category discovery questions like "best X for Y," with the other 98% going to third parties such as listicles, YouTube and other ven…
China's advertising industry issued a five-part standard on generative engine optimization (GEO) between September 9 and 14, 2026, roughly six months after CCTV's March 15, 2026 "3·15 Gala" exposed GEO vendors fabricatin…
A developer has published factfile.md, an open Markdown standard placed at a site's root (yoursite.com/factfile.md) that gives AI agents a business's key facts in a fixed structure, each linked to the page that states it…
NVIDIA reports that its 1B NVFP4 embedding model processes 27.1% more documents than BF16 on GB200 hardware, and that adding its reranker cut inference token use by 13% to 14% in agent simulations while improving retriev…
Tencent opened Yuanbao Search to external developers and businesses through Tencent Cloud on October 20, turning the search infrastructure behind its own AI products into a commercial service for large language models an…
OpenAI unveiled its always-on agent dots at its developer conference in San Francisco on September 29, 2026, with CEO Sam Altman announcing the product, which connects to more than 4,000 applications and runs on its own …
Elastic reports that moving Jina AI models to NVIDIA RTX Pro 6000 Blackwell Server Edition GPUs made the slowest embedding requests up to 8x faster, reranking more than 4x faster, and OCR document-page conversion about 3…
Developer Minh Dao released Fortran MCP, an open-source MIT-licensed local Model Context Protocol server that gives AI coding assistants a curated Fortran documentation corpus covering the Fortran Package Manager, Fortra…
A researcher found that Google Search served random AI Overviews for invalid Knowledge Graph Machine IDs (kgmids), with each page refresh returning unrelated content such as a maths homework answer, commentary on Mid-Cen…
Metehan Yesilyurt, a GEO researcher at Peec AI, published an account of a $50 experiment in which StateGlobe, a website he built with 50,000 fabricated statistics, continued to draw ChatGPT retrievals and citations after…
Miro nearly doubled its analytics agent's accuracy from close to 50 percent to around 90 percent after layering DataHub Cloud in as a governed context platform between Claude Code and Snowflake, according to a DataHub ca…
Luke Wroblewski's Ask LukeW personal AI system now automatically checks for new or changed content every hour, replacing a manual content-ingestion pipeline that previously had to be kicked off for each new article, pres…
Shopify Engineering detailed a cross-store product clustering system that groups listings of the same real-world item across millions of stores under a single Universal Product Identifier (UPID) in the Shopify Catalog AP…