Intercognix aims to surface discoveries between disciplines

Jay Gontu has introduced Intercognix, a concept focused on finding candidate insights that emerge between established fields of knowledge. The effort uses AI to compare large bodies of text and flag relationships that may deserve expert review, testing and validation.

Why it matters: - Intercognix is designed to help researchers spot possible discoveries that sit between disciplines, not just inside them. - The approach matters because important relationships can remain hidden when specialties stay isolated from one another. - AI now makes it practical to compare very large bodies of knowledge across fields at a scale that would be difficult to manage manually.

What happened: - Chicago-area technology professional and author Jay Gontu introduced Intercognix as a framework for identifying candidate insights from previously unrecognized relationships among independently known things. - Gontu publicly presented the concept as a new frontier for discovery while continuing work on how the frontier can be explored systematically. - The concept and related work are available at Intercognix.com.

The details: - Intercognix combines "inter," meaning between, "cogni," referring to knowing, and "X," representing the unknown. - Gontu defines an Intercognix candidate as an insight that can be discovered through previously unrecognized relationships among independently known things. - A computationally identified relationship is not treated as a discovery until it passes prior-art review, domain-expert scrutiny and, where appropriate, empirical testing. - The concept draws on historical precedent from information scientist Don R. Swanson, who in 1986 connected separate medical literatures on fish oil and Raynaud’s syndrome and described the broader phenomenon as "undiscovered public knowledge." - Intercognix extends that idea into the AI era by using systems that can represent and compare concepts across large bodies of text. - Gontu said, "Every question we ask a machine is limited by what we can already imagine. We have built extraordinary systems for answering questions. The harder frontier is discovering the questions nobody has thought to ask — and that is where these unrecognized connections may be hiding." - Gontu also said, "The science on each side is often already done. What remains is the bridge — and building a bridge is a smaller ask than building both shores." - The broader goal is to encourage researchers, students, technologists and domain experts to look beyond traditional disciplinary boundaries.

Between the lines: - The pitch is less about replacing expert judgment and more about expanding the set of hypotheses that experts can evaluate. - The "candidate" framing matters because it sets a high bar for validation and avoids treating machine-found links as finished science. - The idea also reflects a shift in AI use from answering known questions to helping people find questions worth asking.

What's next: - Gontu says the work around systematically exploring this frontier will keep developing. - The next step is likely deeper testing of which cross-disciplinary connections are genuinely useful, reproducible and worth pursuing. - More information and updates are available from Intercognix and Jay Gontu.

The bottom line: - Intercognix is trying to turn AI into a bridge between fields, with the goal of surfacing discoveries that may already exist in plain sight but have never been connected.