Investor Answer Index

The Investor Answer Index is a Fortune 500 AI answers study that measures what ChatGPT, Google AI Overviews, Google AI Mode and Microsoft Copilot tell investors during earnings season. MS House checks each verifiable answer against SEC EDGAR filings. The study will compare accuracy, outdated answers, guidance accuracy and cited sources. No Index results are available yet.

What will the Index measure?

For each company in scope, the Index will report:

• Accuracy rate: the share of checkable AI answers that match the company’s SEC filing.
• Outdated rate: the share of answers that give a figure from an earlier quarter.
• Guidance accuracy: whether engines report the latest guidance, not the prior one.
• Engine scores: accuracy for each engine: ChatGPT, Google AI Overviews, Google AI Mode and Microsoft Copilot.

How should readers interpret an accuracy result?

An Investor Answer Index accuracy result is the share of verifiable answers that match the company’s SEC filings. A 100% result means every answer in that sample matched the filings. It does not mean every AI answer about the company is correct.

Readers should check the sample size, capture date and engine beside each result. A small sample offers less evidence than a large one. Capture dates matter because answers can change. Engine scores remain separate because ChatGPT, Google AI Overviews, Google AI Mode and Microsoft Copilot may give different answers.

MS House does not publish the audit questions. See the methodology and the Acme Inc. sample audit for more context.

How will the Index be built?

MS House will build the Investor Answer Index from company-specific investor questions, captured AI answers and applicable SEC filings. The method keeps each published result tied to a dated answer and a primary source.

How does MS House choose investor questions?

The Investor Answer Index uses a company-specific question set instead of one fixed list. A prediction model draws on question patterns from Google, Bing, DuckDuckGo, ChatGPT, Reddit and Quora. Academic studies of investor search behavior also inform the set.

MS House updates each company’s set every quarter to reflect likely investor interests. Questions about stable company facts, such as the CEO’s identity, may remain. MS House does not publish the exact question wording.

How does MS House audit AI answers against SEC filings?

MS House captures answers from ChatGPT, Google AI Overviews, Google AI Mode and Microsoft Copilot during earnings season. MS House checks each verifiable answer against applicable 8-K Item 2.02 exhibits, 10-Q filings and 10-K filings on SEC EDGAR. For each answer, MS House records the question date, answer capture date, filing date, filing form and filing link. The methodology page explains the checking rules.

How should IR teams measure investor perception in AI search engines?

Investor relations teams should use company-specific topics, capture each engine on the same date and check every verifiable claim against applicable SEC filings. Before an investor day, the review can cover strategy, segments and capital return without publishing the exact question wording.

Teams should keep results separate for ChatGPT, Google AI Overviews, Google AI Mode and Microsoft Copilot. Each result should include the answer date, cited source, filing form, filing date and filing link. Accuracy and outdated answers should be reported separately. This method shows what investors may hear and which filing supports the correct information.

Why do cited source domains matter?

A cited source domain identifies a website that an AI engine presents as support for an investor answer. Checking the domain helps IR teams see whether ChatGPT, Google AI Overviews, Google AI Mode or Microsoft Copilot points to SEC EDGAR, the company’s IR site or a third-party page.

A correct figure may still point to a secondary source. A wrong figure’s citation can identify a page that needs review. The measured study on ChatGPT financial data citation sources reports which domains appeared in captured citations.

What Index data is available now?

No current Investor Answer Index results or downloadable CSV are available on this page. MS House will publish a downloadable CSV with dates and sample sizes after collection is complete. Each published result will show its sample size. Until then, readers can review the Acme Inc. sample audit and why AI chatbots get earnings wrong.

Why build an Index?

The Investor Answer Index will give investor relations teams a benchmark for how AI engines answer investor questions during earnings season. A single company audit shows one result. The Index will show whether that result is normal for the sector or an outlier.

Is there research on Fortune 500 AI answers?

The Investor Answer Index is MS House’s public research on how accurately AI engines answer investor questions about Fortune 500 companies. No current results are available yet. This page therefore does not state how often engines get earnings figures wrong or which errors are most common. After collection, the Index will report accuracy, outdated answers, guidance accuracy and separate results for ChatGPT, Google AI Overviews, Google AI Mode and Microsoft Copilot.

How is the Index different from a company audit?

The Index provides an earnings season benchmark across Fortune 500 companies. The Investor Answer Audit checks what AI engines tell investors about one company and compares each verifiable answer with that company’s SEC filings. The Acme Inc. sample audit shows the report format. Read why AI chatbots get earnings wrong for the common causes.

How can we see our own result now?

You do not need to wait for the Index. The Acme Inc. sample audit shows what a full check of one release looks like, and the Quarterly Audit runs it on your next release.