Working Atopia™ sensors.
Working demo of The Nexus™.
13 patents filed.

A new safety layer for AI.

Today’s AI is powerful, but it has emotional and contextual blind spots.

That’s why regulators and standards bodies are already trying to define “safe” AI, from the NIST AI Risk Management Framework to UNESCO’s Recommendation on the Ethics of Artificial Intelligence. Yet even with those efforts, current systems still don’t sense how options will impact people emotionally before they choose responses, so rule-based guardrails can’t always minimize harm.

What’s missing is Quantified Empathy: using human emotional responses from biosignals to guide ethical AI behavior — paired with a measurable pre-deployment safety score (E³T).

Atopia design evolution

How it Works

The Nexus – Data Engine

The Nexus is a game… but it’s also more than a game. In practice, it’s an immersive interactive experience designed to elicit emotional reactions in carefully designed Scenarios. During gameplay, we collect emotion-related data such as muscle tension, heart rate, heart rate variability, posture, movement patterns, facial expressions, etc. To accomplish this, we designed state-of-the-art Atopia wearable sensors. In addition, these wearable sensors can work in conjunction with passive data from eyes, facial expressions, etc. The Nexus is designed to feed a living dataset: millions of text-to-emotion mappings that allow AI to predict the likely human reaction to any situation.

EmPath – The Affect Layer

From there, we derive Affective State Vectors (ASVs) and convert them into Probable Emotion States (PESs). How accurate is a “Probable Emotional State”? Together, a Scenario, psychophysiological data, ASVs, and PESs form a Node in the structured dataset that trains EmPath. We map relationships between emotional states as Pathways. Then, when a user queries EmPath, it will search the database for the closest Node match, and evaluate Pathways to provide policy-conditioning inputs to guide the model’s responses. This is quantified empathy, the cornerstone of ethically aligned behavior.

We are designing EmPath as a drop-in layer that consumes the Nexus-built dataset. We plan to build it out post-acquisition.

E³T — Safety Standard

Finally, we propose E³T as a measurable framework to evaluate models across multiple dimensions.

Empathy, Emotional Comprehension, and Ethics are quantified as a pre-deployment behavioral safety score.