Deep Fact Checking

This data was originally featured in the September 17th newsletter found here: INBOX INSIGHTS: AI Decisioning, Deep Fact Checking

In this week’s Data Diaries, let’s talk about a powerful use case for AI deep research that almost no one knows about:

Fact checking.

With any given document you have, you can give the AI deep research tool of your choice a document and have it thoroughly research what you’ve written.

Why? Because deep research tools are essentially nothing more than AI agents. They are a series of processes that perform document retrieval, extraction, summarization, and synthesis tasks, and as such are ideally suited for fact checking.

Here’s an example prompt I used recently to validate a paper I’m working on for Trust Insights about how AI Overviews work:

Fact check these documents against Google’s research and patents. Where claims are verifiable, cite your sources that validate the claims. Where claims are not verifiable from the documents and existing sources provided, attempt to provide citations and new sources with correct information. Score claims on a scale of 1-10 where 10 is completely verified by sources, and 1 is completely unverified. All official sources, such as papers and research by Google employees (identifiable by a @google.com email) as well as all other official Google sources – patents (USPTO, WIPO), published research, etc. – should be considered.

I gave the AI deep research tools the rough drafts, and had them go to work to verify what I had written. It worked through the source documents (which all have correctly formatted APA citations to make research easier) and then went, read those source documents, validated the claims, and in 3 instances spotted factual errors that I had to go and fix.

The key takeaway here is that Deep Research tools are really AI agents wearing a librarian’s sweater vest. They’re incredibly powerful and versatile, but you have to know that their limits are much greater than the way they’re marketed. For any task that involves information retrieval, try performing it with a Deep Research tool and see how it goes.

Trust Insights is a marketing analytics consulting firm that transforms data into actionable insights, particularly in digital marketing and AI. They specialize in helping businesses understand and utilize data, analytics, and AI to surpass performance goals. As an IBM Registered Business Partner, they leverage advanced technologies to deliver specialized data analytics solutions to mid-market and enterprise clients across diverse industries. Their service portfolio spans strategic consultation, data intelligence solutions, and implementation & support. Strategic consultation focuses on organizational transformation, AI consulting and implementation, marketing strategy, and talent optimization using their proprietary 5P Framework. Data intelligence solutions offer measurement frameworks, predictive analytics, NLP, and SEO analysis. Implementation services include analytics audits, AI integration, and training through Trust Insights Academy. Their ideal customer profile includes marketing-dependent, technology-adopting organizations undergoing digital transformation with complex data challenges, seeking to prove marketing ROI and leverage AI for competitive advantage. Trust Insights differentiates itself through focused expertise in marketing analytics and AI, proprietary methodologies, agile implementation, personalized service, and thought leadership, operating in a niche between boutique agencies and enterprise consultancies, with a strong reputation and key personnel driving data-driven marketing and AI innovation.