LoginSign Up
← Back to all sparks
M

Microsoft Bing

AI-ASSISTANTS
Velocity4.3

Microsoft search engine with AI-powered Copilot, image search, and webmaster tools.

www.bing.com ↗

Bing pivots from ranking pages to grounding AI, shipping APIs and an open embedding model

groundingai-searchembeddingswebmaster-toolscitation-analyticsagentic-apis
◆Current state
Bing is repositioning its search index as the grounding layer for AI assistants rather than a destination for human browsing. Recent shipping reflects this: Web IQ grounding APIs, an open-source embedding model topping MTEB-v2, and AI-citation reporting for publishers in Webmaster Tools. The consumer-facing image-search refresh is the exception in an otherwise infrastructure-and-publisher-tooling agenda.
◆Where it's heading
The throughline across entries is grounding: feeding fresh, verifiable web data to agents and assistants, then giving publishers visibility into how their content gets cited. Bing is building the supply side (APIs, embeddings) and the measurement side (citation share, intents, topics) of the AI-answer economy simultaneously. The framing essays signal Microsoft intends to own grounding as a category.
◆Prediction
Expect the Webmaster Tools AI-visibility previews to reach GA and Web IQ to add pricing tiers or expanded data types as it courts third-party agent builders.

◆Recent moves

  1. 20d ago

    New AI Visibility Insights in Bing Webmaster Tools: Intents, Topics, Citation Share, Compare

    Four new reports (Intents, Topics, Citation Share, Compare) extend AI Performance in Webmaster Tools, giving publishers deeper read on how AI answers cite them. Real new tooling, but an additive layer on the existing AI Performance preview.

    View source ↗
  2. 1mo ago

    Announcing Microsoft Web IQ

    ⚡ SPARK

    Web IQ packages Bing's grounding into a suite of APIs for AI agents, marking Bing's move from search destination to grounding-as-a-service. It is the clearest expression of the index-to-grounding pivot the feed has been building toward.

    View source ↗
  3. 1mo ago

    A Smarter Way to Explore Images Has Come to Bing

    An AI-guided reorganization of image search results aimed at reducing overload. A visible consumer UX change, incremental against Bing's larger grounding agenda.

    View source ↗
  4. 1mo ago

    Keeping Trusted Content Visible in an AI-Powered Search World

    A positioning post on safety-by-design and trusted-content prioritization in AI search. Framing and policy narrative, no specific shipped capability.

    View source ↗
  5. 2mo ago

    Evolving role of the index: From ranking pages to supporting answers

    A thought-leadership essay on the index shifting from ranking pages to supporting AI answers. Strategy narrative that frames the grounding pivot but ships nothing.

    View source ↗
  6. 3mo ago

    Microsoft Open-Sources Industry-Leading Embedding Model

    ⚡ SPARK

    Harrier is an open-source embedding model series ranking first on multilingual MTEB-v2, underpinning Bing's grounding quality. Open-sourcing a SOTA model is a directional bet to make Bing's grounding stack the default substrate for AI systems.

    View source ↗