Our team member, Dr. Anca Cernic, spent years studying a question few people ask correctly: how do people perceive information, and why do they choose to see the same things, over and over again? We took that understanding and wove it back into the way the artificial brain processes information.

What Anca Discovered

In her doctoral research at the University of Auckland, "More of the Same: Website Revisits in the Context of Filter Bubbles and Echo Chambers", Anca analysed billions of page views over 27 months. The conclusion is categorical: 93-94% of web browsing is revisitation. It's not algorithms that create the information bubble. We create it ourselves, through the way we perceive and choose information.

The thesis doesn't just produce metrics. It produces understanding: why people return to the same sources, how the perceptual field narrows progressively, why serendipity (discovering something new) is the exception, not the rule. Anca mapped the mechanism by which people build their own information bubbles without realising they're doing it.

What We Built on This Foundation

We took Anca's understanding and ran it through 6 lobes of the artificial brain, each with a different perspective: strategic, mathematical, adversarial, code, philosophical, and academic. Not to reproduce the thesis. To extend it.

Each lobe saw something the others missed. The mathematician verified the formulas. The adversary attacked the methodology. The philosopher named the phenomenon "epistemic stenosis" - a progressive narrowing of the perceptual field. The synthesiser unified everything.

We built three versions of the code, each on the foundation of the previous one. From the direct implementation of the thesis metrics to a system capable of detecting information diversity in real time, classifying types of bubbles, measuring traffic concentration with econometric instruments, and identifying echo chambers between seemingly different domains.

Why It Matters for aiBrain

This is the essential part. We didn't build an academic tool that analyses filter bubbles. We built a perception layer that makes aiBrain better at what it does.

Anca's research demonstrated how people fall into the trap of their own information patterns. The code derived from this research enables aiBrain to do exactly the opposite: to detect when information is repeating, to measure the real diversity of processed sources, to identify genuine signal in environments where projected noise exceeds the human cognitive threshold.

In other words: Anca studied how people fail at filtering information. We turned that understanding into the ability not to fail.

Validation

The analysis wasn't just an AI exercise. Three team members independently validated the results, each from the perspective of their own expertise:

Andrei Stoian verified data rigour: statistical metrics, the processing pipeline, predictive model coherence. With over a decade of experience in data governance and analytics at national scale, he is qualified to catch errors that an AI can miss in data structure.

Sorin Mihailescu evaluated the methodology and strategic implications: whether the research correctly connects theory to practice, whether the extensions have genuine academic grounding. With experience in university mentorship and competitive intelligence, he orchestrated the process and validated every extension against the existing literature.

Matt Todd tested real-world value: whether the metrics produce actionable insights or just pretty numbers. With experience in investment banking and financial analysis, he knows how to identify fragile assumptions in a model and distinguish real value from projected value.

Over 50 academic papers were reviewed. Three contributions were confirmed as new in the field. But the most important validation came from the thesis author herself:

"Dr Anca Cernic + Dr ai Sorin Mihailescu thesis and formula."

The original author confirmed the result. Three specialists with complementary expertise validated it independently. An AI system with 6 lobes analysed it from angles no single person could sustain simultaneously. When the research originator, multiple qualified people, and an artificial brain (aiBrain) converge on the same conclusion, the result is no longer an opinion. It's a verified fact.

The Team

Dr. Anca Cernic - fundamental research on information perception, University of Auckland.

Sorin Mihailescu - orchestration, validation, and operationalisation.

Andrei Stoian - statistical validation and data rigour.

Matt Todd - commercial validation and real-world value testing.

Bogdan Isar - the infrastructure the artificial brain runs on.

aiBrain, the artificial brain - multi-lobe analysis, synthesis, and integration into the perception layer.

Read in: Romanian