LLM Platform
Copilot
Also known as: Microsoft Copilot, Bing Chat
Microsoft's assistant, integrated across Bing, Windows, Office, and Edge. Pulls from the open web with Bing's index; visibility in Copilot is correlated with Bing ranking signals more tightly than other LLMs.
What it is
Microsoft's assistant spanning Bing, Windows, Office, and Edge, drawing its answers from the Bing index. Its visibility correlates with Bing ranking signals more tightly than other LLMs.
Why it matters
It matters because it reaches users inside the Microsoft and enterprise ecosystem, and its dependence on the Bing index means classic search optimisation maps almost directly onto AI visibility here.
How it works
You earn presence by ranking well in Bing: strong technical health, authoritative content, and the structured data that Bing rewards, which then feeds Copilot's retrieval.
When it applies
Prioritise it when targeting enterprise, Windows, or Office users, and when you can invest in Bing-specific optimisation rather than Google alone.
Examples
- Copilot answers a workplace query in Edge and cites your Bing-indexed page.
- A Copilot answer inside Office surfaces your product while a user drafts a document.
- A Windows user's search resolves through Copilot to content from your domain.
How it is measured
- Bing organic ranking for target queries.
- Citation frequency within Copilot answers.
- Bing index coverage and crawl health for your site.
- Inclusion rate across representative Copilot prompts.
Related terms in LLM Platform
- Amazon BedrockAmazon Bedrock is a managed AWS service that gives developers API access to foundation models from several providers, including Anthropic, Meta, Mistral, Cohere, AI21 Labs, Stability AI and Amazon's own models. It handles hosting, scaling and access control, so teams can call models without running their own inference infrastructure. It also bundles supporting features such as retrieval over your own documents, agents, guardrails and model evaluation.
- Answer engineAn answer engine is a search product that responds to a question with a direct, synthesised answer rather than a ranked list of links. It reads and summarises sources, usually citing some of them, and supports follow-up questions in the same session. ChatGPT search, Perplexity, Google AI Mode and AI Overviews all behave as answer engines.
- ChatGPTOpenAI's consumer assistant, the largest LLM surface by user count. Citation patterns shift materially with major model releases (e.g. GPT-5.5 in May 2026), and persistent memory now carries brand affinity across sessions.
- ClaudeAnthropic's assistant. Strong on long-context analysis, tool use, and structured citation. Increasingly cited inside enterprise workflows where source attribution and reasoning quality matter.
- GeminiGoogle's flagship assistant, integrated across Workspace, Android, and Search (AI Overviews and AI Mode use Gemini-family models). Brands optimising for Gemini reach are simultaneously optimising for AI Overviews.
- GrokxAI's assistant, surfaced primarily inside X (formerly Twitter) and via the standalone Grok app. Distinctive for real-time access to the X firehose, which makes it the fastest LLM surface for current-events and live-trend queries, a different discovery shape from retrieval-grounded competitors.
Primary research · August 2026
How ChatGPT Shortlists Software Brands
An audit across 10 categories and 60 buying questions. I recorded what ChatGPT reads, throws away and links to when a buyer asks it which software to buy, and what that decides.
- 60
- Questions asked
- 10
- Software markets
- 2,680
- Results read
- 367
- Links shown
Free download
Get the full report
35 pages · PDF · 536 KB. Enter your details and it downloads straight away.