Brain-like computing: Thinking outside of the chip

• Author:
• Author name

  Quantumrun Foresight
• August 14, 2024

Insight summary

Brain-like computing mimics the human brain's structure to improve efficiency and reduce energy use in technology. Researchers have developed systems that can perform complex tasks with high accuracy and significantly lower power consumption than traditional models. This trend can transform personal technology, business operations, and public services, making them more adaptive and efficient.

Brain-like computing context

Brain-like computing, also known as neuromorphic computing, is an innovative approach that emulates the structure and functionality of the human brain to enhance computational efficiency. This field aims to replicate the brain's architecture using spiking neural networks and advanced materials like nanowires. Traditional computer architectures face significant challenges in processing and storing vast data. In contrast, brain-like computing offers a promising alternative by integrating memory and processing capabilities within a single system, much like the human brain.

One notable advancement comes from researchers at the California NanoSystems Institute at UCLA. In November 2023, they developed an experimental system that physically models the biological brain using a network of silver nanowires. This system demonstrated an impressive 93.4 percent accuracy in identifying handwritten numbers, outperforming conventional machine-learning approaches. The success of this system highlights the potential of neuromorphic computing to perform complex tasks with significantly lower energy consumption compared to traditional silicon-based AI systems. This research underscores the importance of co-designing hardware and software to optimize performance.

Meanwhile, researchers at the University of California, Santa Cruz developed SpikeGPT in 2023, the largest spiking neural network designed for language generation. SpikeGPT consumes 22 times less energy than conventional deep learning models while maintaining high performance, showcasing the efficiency of neuromorphic approaches. This system leverages the brain's temporal dimension, storing information in the timing of neural spikes, a fundamental difference from traditional AI models that continuously consume power. As researchers continue to explore brain-like computing, the field promises to revolutionize areas such as robotics, autonomous navigation, and the Internet of Things by providing energy-efficient, adaptive systems capable of processing complex data in real time.

Disruptive impact

Brain-like computing can enhance personal technology and daily life. For example, devices such as smartphones, wearables, and home assistants may become far more efficient, capable of learning user behaviors and preferences in real-time with minimal power consumption. Health monitoring systems could benefit significantly by providing continuous, adaptive feedback based on biometric data. This trend can improve personal productivity and well-being by offering more intuitive, responsive, and reliable technology.

Companies may adopt neuromorphic systems to process large datasets quickly and efficiently, reducing operational costs and energy use. This technology could enhance manufacturing, logistics, and customer service automation by enabling machines to learn and adapt to new tasks on the fly. Firms could also innovate new services and products, such as adaptive AI-powered applications and devices, creating competitive advantages in the marketplace.

Meanwhile, policymakers may need to create frameworks to regulate the ethical use and security of these advanced systems. There is potential for significant benefits in public services, such as smarter infrastructure, efficient traffic management, and enhanced national security systems. Countries investing in neuromorphic research may gain strategic advantages, but there could also be an arms race in AI technology, requiring careful diplomatic efforts to manage global AI policies and prevent misuse.

Implications of brain-like computing

Wider implications of brain-like computing may include: 

• A rise in adaptive learning technologies leading to more personalized and effective education.


• Increased energy efficiency in data centers reducing environmental impact and operational costs.


• A shift towards more automated workplaces reducing the need for repetitive manual labor while increasing demand for skilled technicians.


• Enhanced capabilities in healthcare diagnostics and monitoring leading to earlier disease detection and better patient outcomes.


• New ethical guidelines and regulations developed to ensure the responsible use of brain-computer technologies.


• Advanced brain-like computing becoming integral to smart city infrastructure improving urban planning and resource management.


• Greater economic disparity as companies with access to advanced AI outpace those without, widening the technology gap.


• Reduced reliance on centralized cloud services as neuromorphic computing enables more effective edge computing solutions.


• A potential decrease in cybersecurity risks as adaptive AI systems identify and mitigate threats more effectively.


• Demographic shifts in the workforce as more people pursue careers in brain-computer technologies and related fields, altering educational priorities.

Questions to consider

• How could brain-like computing enhance the technology you use daily, such as smartphones and home assistants?


• What potential benefits and challenges might your workplace face with adopting brain-like computing systems?