Tools

Microsoft launches Web IQ: Bing-powered grounding API for AI agents

Updated: View as Markdown RSS

Microsoft has announced Web IQ, a suite of AI-native grounding APIs built on Bing’s search index and designed specifically for how AI agents retrieve information. The announcement itself, made at Microsoft Build 2026, names no customers. Speaking to Search Engine Land, Jordi Ribas, Microsoft’s President of Search and AI, said Web IQ “is used by Bing to show the Copilot answers at the top of the Bing search results for some queries, it is used by ChatGPT for some of its web answers and it is used directly in Copilot for answers”. The qualifier matters: this is a substantial share of ChatGPT’s web retrieval, not the whole of it.

Where Bing was built for human searchers, Web IQ is built for AI inference: retrieving the right passages, packaging them as structured evidence, and returning them within the latency budgets that multi-step agentic reasoning requires.

What Web IQ does differently

Traditional search returns ranked documents. Web IQ also returns passages and structured evidence objects: discrete extracts from relevant pages, selected for their information density and assembled into structured context for the requesting model. This reduces the number of tokens a model needs to process and eliminates the step of parsing full-page HTML.

Microsoft measures the quality of Web IQ’s grounding with GDSAT (grounding satisfaction), a metric that “captures whether the grounding truly meets user intent across completeness, freshness, and authority”. It is how Microsoft grades its own system’s results against alternatives; Microsoft publishes no rule for how an individual page or passage is scored.

Correction, 28 September 2026: An earlier version of this article said GDSAT determines which passages surface in a response, and that it applies at passage level rather than page or domain level. Microsoft’s announcement describes GDSAT as a measure of Web IQ’s results, not a selection rule.

Performance claims: sub-165ms p95 latency across five data centres, and roughly 2.5× faster than the next alternative in comparable configurations.

How Web IQ works

Microsoft rebuilt its retrieval stack from indexing through to orchestration to meet the demands of agentic workloads. Key components:

  • DiskANN-based vector search for large-scale approximate nearest-neighbour indexing, extended from Microsoft’s open-sourced embedding research
  • Fan-out orchestration layer that interprets an agent’s sub-query, retrieves candidate passages across multiple sources, and ranks and assembles results as structured evidence
  • Content understanding models trained specifically for LLM-driven reasoning rather than human relevance signals
  • Passage selection that maximises information density while minimising token cost per call

Jordi Ribas, President of Search and AI at Microsoft, described the shift: “Agents want to extract the right information from documents, package it and deliver it quickly.”

Publisher controls

Web IQ inherits Bing’s existing robots.txt compliance and publisher preferences. Microsoft has stated it is engaging with standards bodies, including the IETF, on interoperable frameworks for publisher rights in the AI era.

No new crawler user-agent is introduced by Web IQ. It draws from Bing’s index, so existing BingBot configurations govern what Web IQ can access.

What this means

Some of ChatGPT’s web retrieval can be passage-level. The written announcement names no customers, but Ribas has said on the record that ChatGPT uses Web IQ for some of its web answers. To the extent that content reaches ChatGPT Search responses through this route, it can reach the model as a passage rather than as the whole page.

GDSAT names what Microsoft grades its grounding on. Completeness, freshness and authority are its three dimensions. They align closely with GEO guidance on direct answers, recency and credible sourcing, but Microsoft uses them to measure its system’s results, not as published ranking signals.

No new optimisation actions are required. Web IQ runs on BingBot’s existing crawl. Publisher controls work through robots.txt as usual.

Access to Web IQ for third-party developers remains limited: Microsoft is accepting expressions of interest at webiq.microsoft.ai but has not announced general availability or pricing.

Sources

More news

Add as a preferred source on Google