Despite remarkable advancements in artificial intelligence, the human brain continues to outperform machines as the world's most adaptable and efficient information processor. While humans can rapidly make decisions with incomplete or evolving information, many contemporary AI systems require extensive training on precisely labeled data and demand complete retraining when new information becomes available.
Enter Nara Logics, an innovative startup co-founded by an MIT graduate, which is pioneering the next frontier of brain-inspired artificial intelligence technology. The company's groundbreaking AI engine leverages cutting-edge neuroscience research to replicate the brain's structure and functionality at the neural circuit level, creating a system that more closely mirrors human cognition than ever before.
The result is an advanced AI platform offering numerous advantages over conventional neural network systems. Unlike traditional systems that rely on meticulously tuned, fixed algorithms, users can actively engage with Nara Logics' platform, adjusting variables and objectives to explore their data more thoroughly. This neuroscience-based AI decision making system can operate without labeled training data and seamlessly integrate new datasets as they become available. Most significantly, the platform provides transparent explanations for every recommendation it generates—a crucial feature driving adoption in critical sectors like healthcare.
"Many of our healthcare clients have used AI systems that predict the likelihood of patient readmission, but they've never received the 'but why?' explanations needed to take meaningful action," explains Jana Eggers, CEO of Nara Logics, who leads the company alongside CTO and founder Nathan Wilson PhD '05.
Currently, Nara Logics' explainable AI for healthcare applications is being deployed by healthcare organizations, consumer companies, manufacturers, and government agencies to reduce costs and enhance customer engagement strategies.
"Our solution serves professionals facing increasingly complex decisions due to growing data volumes and factors, as well as those approaching complex decisions differently because novel information has become available," Eggers notes.
The platform's revolutionary architecture stems from Wilson's commitment to embracing neuroscience's complexities rather than oversimplifying them. He developed this approach during more than a decade working in MIT's Department of Brain and Cognitive Sciences, which has long focused on reverse-engineering the human mind.
"At Nara Logics, we believe neuroscience is on a trajectory that will lead to groundbreaking decision-making approaches we haven't seen before," Wilson states.
Pursuing a Vision
Wilson completed his undergraduate and master's degrees at Cornell University, but after arriving at MIT in 2000, he remained dedicated to the institution. Throughout a five-year PhD and seven-year postdoc, he developed mathematical frameworks to simulate brain function.
"The MIT community is intensely focused on developing new computational models that transcend traditional computer science," Wilson explains. "This work intersects with computer science while also exploring how our brain's processes could inform computer functionality or inspire new computing paradigms."
During nights and weekends in the final years of his postdoc (2010-2012), Wilson began transforming his algorithms into a commercial system that would eventually become Nara Logics' foundation. In 2014, his work attracted Eggers' attention—despite her successful business background, she had grown skeptical of AI hype.
Eggers became convinced that Nara Logics' AI engine represented a superior approach to helping businesses. Even then, the engine—now called Nara Logics Synaptic Intelligence—possessed unique properties that set it apart in the field.
Within the engine, objects in customer data (such as patients and treatments) self-organize into matrices based on shared features, creating a structure resembling biological systems. Relationships between objects form through local functions the company calls synaptic learning rules, adapted from neuroscience research at cellular and circuit levels.
"Our process catalogs all metadata and what we term our Connectomes mine unstructured data databases, establishing connections across all related elements," Wilson elaborates. "Once this foundation exists, users can specify preferences and let the engine analyze the data to find matches to those parameters. The crucial advantage is eliminating the need to predefine 'correct answers' based on similar cases—bypassing that entire step."
Each object within Nara Logics' Synaptic Intelligence stores its properties and rules locally, enabling the platform to adapt to new data by updating only a small number of associated objects—a bottom-up approach believed to mirror how the brain operates.
"This differs fundamentally from deep learning or other approaches that globally optimize everything, with each cell following the global algorithm's directives," Wilson clarifies. "Neuroscience research indicates that each cell makes autonomous decisions to some extent."
This design allows users to explore data relationships by 'activating' specific objects or features and observing what else becomes activated or suppressed in response.
When providing answers, Nara Logics' engine activates only a small subset of objects in its dataset. The company compares this to the 'sparse coding' principle believed to operate in higher brain regions, where only a limited number of neurons fire at any given moment. This sparse coding approach enables the company to trace its platform's reasoning pathway and provide users with explanations for its decisions.
As the company has evolved, Wilson has maintained connections with MIT's research community, and Nara Logics participated in the STEX25 startup accelerator, operated by the MIT Startup Exchange.
Harnessing Human-Like Artificial Intelligence Solutions
Manufacturers are already leveraging Nara Logics' platform to better understand IoT device data, consumer companies are using it to strengthen customer connections, and healthcare organizations are applying it to improve treatment decisions.
"We're focused on a specific algorithm—the mechanics of decision-making," Wilson emphasizes. "We believe this process can be codified and that getting it right will deliver extraordinary value."
As COVID-19 disrupted industries and highlighted the need for adaptive software tools, Nara Logics nearly doubled its customer base. The founders are excited about scaling a solution they believe is more collaborative and responsive to human needs than other AI systems.
"We believe our most significant contribution is developing an AI where humans actively participate and remain in the loop—they're conscious, understanding, and aware of what the system is doing," Wilson concludes. "This enables smarter daily decisions that accumulate to create substantial impact over time."
