IntuiCell augments off the shelf quadruped with a ‘digital nervous system’

IntuiCell, a deep-tech startup that aims to blend neuroscience and artificial intelligence, this week said it has developed a system capable of real-time learning in real-world settings. The Stockholm-based company released a video of an off-the-shelf quadruped robot — with no pre-programmed intelligence or instructions — that it has augmented with a “digital nervous system.”

In other words, IntuiCell created a physical AI agent capable of autonomously learning in a manner that mimics a biological nervous system. The company, a spin-out of Sweden’s Lund University, claimed that its technology moves beyond static machine-learning models, the mainstay of traditional AI, to a fully functional system capable of scaling to human-level intelligence.

“For decades, AI has excelled at processing vast amounts of data but has fundamentally been incapable of real intelligence,” stated Viktor Luthman, co-founder and CEO of IntuiCell. “Our system changes that. We’ve translated the principles of biological learning into software, enabling AI to evolve, adapt, and interact with the world in ways never previously envisaged.”

“Through licensing our digital nervous system, we aim to become the infrastructure for all non-biological intelligence — empowering others to solve real-world problems we cannot foresee today, without a reliance on massive training datasets,” he said.

IntuiCell said this innovation will enable it to create machines that learn from people, just as a person would teach a new skill to an animal. To further demonstrate this concept in action, IntuiCell plans to hire a dog trainer — rather than integrate a large language model (LLM) and vast datasets — to teach a physical agent new skills.

IntuiCell address current AI model limits

Unlike traditional AI models that are bound by static training data, IntuiCell said its robot dog, named Luna, perceives, processes, and improves itself through direct interaction with its world. The company asserted that Luna demonstrates a functional digital nervous system in action. The robot has learned to control its body and teach itself to stand through trial, error, and cumulative experience.

IntuiCell also said it has the functioning basis for technology that will operate as the Thalamocortex, the part of the brain that processes and predicts the world. It said it expects to complete the full digital nervous system within the next two years.

Ultimately, the company’s goal is to enable the software platform to equip any agent, physical or digital, with lifelong learning and adaptation to the unknown capabilities.

Currently, practically all AI models rely on backpropagation, the training process of feeding error rates back through a neural network. They also use predefined datasets, limiting their ability to learn dynamically, according to IntuiCell.

From LLMs to deep reinforcement learning, as well as some outliers that claim brain-inspired approaches, much of AI is dominated by foundation models trained on vast datasets, it said.

These approaches have achieved results in static, low-stakes applications such as generating text and images, noted IntuiCell. However, they are not yet able to replicate biological intelligence’s ability to solve unseen problems or engage in high-stakes applications such as robotics or industrial automation, it said.

IntuiCell said its digital nervous system enables AI agents to learn from their own experiences without needing preloaded data or backpropagation. It can also adapt in real time to new environments and challenges, as well as scale naturally toward human-level intelligence, explained the company.

A few of the other companies working on advancing AI and robotics include Sanctuary AI, Neura Robotics, Google DeepMind, Boston Dynamics, and Figure AI.

Learn more about AI at the Robotics Summit

The 2025 Robotics Summit & Expo, which will be on April 30 and May 1 at the Boston Convention and Exhibition Center, will feature a track of sessions around AI and robotics:

• “Risk Is Different With AI: Strategies for Resilient Robots,” with SK Gupta, co-founder and chief scientist at GrayMatter Robotics
• “Advancing Robot Learning With Force Sensing and Physical Intelligence,” with Klajd Lika, founder and CEO of Bota Systems
• “From Chaos to Clarity: Using AI to Find Important Events in Robot Logs,” with Benji Barash, co-founder and CEO of Roboto AI
• “Generative AI’s Impact on Robotics,” with
  • Ted Larson, CEO of OLogic
  • Xavier Chi, co-founder of mbodi AI
  • Rajat Bhageria, founder and CEO of Chef Robotics
  • Karthee Madasamy, founder and managing partner at MFV Partners
  • Eugene Demaitre, editorial director of The Robot Report

---

---