AI & TechArtificial IntelligenceBigTech CompaniesDigital MarketingNewswireTechnology

Microsoft launches Web IQ: Bing-powered search for AI agents

▼ Summary

– Microsoft released Web IQ, a suite of AI-native grounding APIs that connect AI systems to real-time web data including pages, news, images, and videos.
– Web IQ is rebuilt from the ground up for efficiency and speed, used by Microsoft Copilot and ChatGPT for web answers.
– Unlike human-focused search, Web IQ prioritizes extracting and delivering relevant information for AI agents over traditional ranking.
– Web IQ is roughly 2.5x faster than alternatives and designed to minimize token usage for cost-effective agent operations.
– Currently used by Copilot and ChatGPT, Microsoft plans to expand access to Web IQ as it scales.

Microsoft has officially unveiled Web IQ, a new grounding API designed to serve the growing ecosystem of AI agents. According to the company, Web IQ is a “suite of AI-native grounding APIs built for the agentic era, connecting AI systems and agents to fresh, real-world intelligence from across the web , including web pages, news, images, and videos.” This marks a significant step in how AI systems access and process live online data.

At its core, Web IQ is powered by Bing’s index and leverages the search engine’s deep understanding of web content. The API is built on the same infrastructure that runs inside Microsoft Copilot and several other major AI platforms, including ChatGPT. However, this is not a recycled version of the APIs used when those large language models first launched. Jordi Ribas, President of Search & AI at Microsoft, explained that Web IQ has been “rebuilt from the ground up to be efficient, fast and of course, relevant.” It is already being used by Bing to surface Copilot answers at the top of search results, by ChatGPT for certain web-based responses, and directly within Copilot itself.

A key distinction is that Web IQ is optimized for AI agents, not human users. Traditional Bing Search prioritizes ranking and presentation for human eyes, but Ribas told Search Engine Land that ranking carries less weight for agents. Instead, agents need to efficiently extract the right information from documents, package it, and deliver it quickly. Unlike humans who often enter a single query and stop, agents tend to search deeper and fan out across multiple sources. Microsoft noted that this required “re-architecting the system from the ground up from indexing and retrieval to ranking, passage selection, and orchestration so every layer is aligned around the needs of inference-time grounding.”

Efficiency is a core design principle. Because agents search so frequently, Web IQ was built to use the fewest tokens possible. Microsoft stated that “fewer tokens in, better answers out, lower cost per call.” The company also claims the API is roughly 2.5 times faster than the next best alternative.

Currently, Web IQ is available to Microsoft for Copilot, OpenAI for ChatGPT, and other large LLM platforms. Microsoft plans to expand access as the system scales. Those interested in learning more can express interest on the dedicated Web IQ page.

For the web industry, this development signals a critical shift. As the internet evolves to support agentic experiences, businesses must watch these changes closely and be ready to adapt. The web is being reshaped for AI consumption, and while human users are not disappearing, AI agents are already arriving. Preparing for this next evolution is no longer optional,it is essential.

(Source: Search Engine Land)

Topics

web iq release 98% ai-native grounding 95% search for agents 92% bing index 90% agentic era 88% copilot integration 86% api infrastructure 85% chatgpt integration 84% re-architecting system 82% token efficiency 80%