OCCAM'S FORENSIC JURY
Home Insights Five Modes of Jury Disagreement
Methodology Essay · Be The Exception · Video

My AI Methodology Has 5 Modes of Disagreement. Here's What Each One Means.

5 different published stock valuations on the same company. 5 different reads. Whose story is your story? Most equity research averages the disagreement away. Mine preserves it because the disagreement itself is often the value compass.

Author: John Gillespie
Organization: InsightfulAgents.AI LLC
Published:
Format: Methodology essay + video, 4:10

Educational publication. Not investment advice. Neither John Gillespie nor InsightfulAgents.AI LLC is a registered investment adviser.

As you probably know, if you look at 5 different published stock valuations, you may get 5 different reads. With Occam's Forensic Jury™, we surface the disagreement and the reasons for it so that you can make sure that you agree with the thesis before you invest your hard-earned money.

If the embed does not load, watch directly on YouTube at youtu.be/JRrbNJJSwVE.

5 valuations, 5 reads. Whose story is your story?

Given 5 different stock valuations on the same company, how do you determine which one is correct? Most don't make all of their assumptions clear, so how would an investor know? How would you choose? Each one of the 5 has arrived at one opinion, but is their story your story? How will you decide?

My methodology makes equity research disagreement transparent and preserves it through 4 lenses: the Auditor, the Architect, the Storyteller, and the Sentinel. Each read processes company data from a different angle. When they agree, the verdict consolidates. When they disagree, the disagreement becomes the value compass to help you navigate choppy investment waters. Most equity research averages disagreement away. My methodology preserves it because the disagreement itself is often the point.

Auditor
Financial statements.
Architect
Management's disclosure language.
Storyteller
What isn't in the narrative.
Sentinel
Everything outside the filing.

5 shapes the disagreement takes

Across cycles, those disagreements have taken five shapes. Each shape tells you something different about what the methodology is doing.

Mode 1
Disagreement across runs — address inputs, not architecture

Mode 1 is when the methodology disagrees across runs; we address inputs, not architecture. It is not uncommon for data inputs to require correction.

Mode 2 MongoDB
A single lens flags what averaging would smooth away

Mode 2 occurs when one lens escalates to FLAGGED on something that doesn't add up. Something the other lenses smoothed over by averaging. The methodology's FLAGGED-override architecture promotes that dissent to the headline verdict instead of averaging it away. This was demonstrated in our MongoDB video. The Architect flagged a gap between the operational health score and the DCF that management's disclosure language never bridged.

Mode 3 Qualcomm
Methodology holds the line while the market moves

Mode 3 occurs when one of the lenses escalates to FLAGGED. The methodology's job isn't to predict where the price goes. The methodology's job is to surface whether the disclosure supports the price. Discipline to NOT walk back the dissent IS the output. This was demonstrated in our Qualcomm video.

Mode 4 NVIDIA
Dissent preserved without firing an override

Mode 4 is similar to Mode 2, but the dissenting lens provides a list of things to listen for in subsequent disclosures as described in our NVDA video.

Mode 5 Early production
Naming the kind of disagreement

Mode 5 is now in early production, with case studies to follow. Now the methodology can name the kind of disagreement. Is it because market conditions explain the gap? Is it because the lenses picked different assumptions? Or is there no disagreement at all? Three different shapes.

What's the question you actually want answered?

If this was useful, please like and subscribe. What's the question you actually want answered about AI and investment performance? Drop it in the comments. Drop a ticker. The next stock we cover gets named there, not here. The production calendar lives with the audience.

Regardless of which mode fires

Regardless of which mode fires:

This is forensic equity research as a methodology-output discipline, not a prediction discipline.

Where this fits

Each mode has a richer walkthrough on the public record:

Full research index →

Related

§ 4. Disclosure and disclaimer

OCCAM'S FORENSIC JURY™ is a trademark of InsightfulAgents.AI LLC. © 2026 InsightfulAgents.AI LLC. All rights reserved. This piece is general and impersonal commentary on the methodology's architecture for educational purposes. It does not constitute personalized investment advice. The methodology produces moment-in-time analytical reads. Outputs are not buy or sell recommendations. Past methodology output does not guarantee future results.

The author and InsightfulAgents.AI LLC are not registered investment advisers, broker-dealers, or financial planners under the Investment Advisers Act of 1940 or any state-level analogue, and nothing in this article should be construed as creating a fiduciary relationship.

The author may or may not hold positions in any security or entity discussed; positions can change without notice. All claims are sourced to SEC filings, earnings transcripts, or named third-party data. Past performance does not guarantee future results.

For a full disclosure framework, see /legal.html.