Neurologyca, a company working on human-centric artificial intelligence, announced the launch of a research division on Tuesday dedicated to advancing AI models’ ability to understand and adapt to people.

The research lab, named Neurologyca Labs, will formalize the scientific development of the company’s Human Context AI platform, which will enable intelligent systems to recognize, interpret and respond to human intent. The idea behind it is to give AI models the ability to “feel” their way through human conversations and understand body language as other humans do.

This would include interpreting cues such as trust, stress, cognitive load, confidence and readiness in real-world situations.

“Much of the industry’s attention has been focused on making AI more capable, which has produced incredible breakthroughs,” said Chief Strategy Officer Marc Fernandez. “The next frontier is making AI more aware of the people it is interacting with.”

The initiative will support researchers at universities, research institutions, enterprise outfits, healthcare organizations, AI platform companies and industry partners working to advance the science of human-aware AI.

The company is best known for developing its own form of “emotionally aware AI” in a multimodal platform called Kopernica. The platform can view live video and discern how a user may feel by analyzing more than 790 reference points on the human body. It can also process audio, capturing patterns of speech such as tone, rhythm and cadence.

Neurologyca said that key focus areas of research will be the development of benchmarks and evaluation methods beyond current measures that target only raw intelligence, performance and coding capabilities. Today, most AI systems are marketed for how well they perform in rigidly defined arenas, such as autonomous and agentic roles, where they take on work that would otherwise be laborious or tedious.

However, some AI systems are designed to interact with humans directly, chat with them and provide an interface to computer systems similar to interacting with another person or coworker. Many of these systems are being given human-like avatars, voices, names and faces. As a result, they need to maintain trust and adapt appropriately when human emotions and needs change.

“Human behavior is not static. During any interaction with an AI system, our intent can shift, our confidence can rise or fall, we can become overloaded, we can hesitate, and our trust can change,” added Fernandez. “If AI is going to become a meaningful partner in decision-making, it needs to understand and respond to those dynamics in real time.”

The company said it plans to publish a collection of technical papers and research reports outlining topics including context-aware AI agents, human-centered robotics, AI alignment, digital wellness and human-machine interaction. The first reports are expected to be released in the coming months and will be available via the Neurologyca Labs research portal.

Image: SiliconANGLE/Microsoft Designer