CyberSwarm is developing a neuromorphic System-on-a-Chip dedicated to cybersecurity which helps organizations secure communication between connected devices and protect critical business assets.
Our patented technology can both secure and perform local processing at the same time, enabling hardware neural networks for a bottom-up cybersecurity approach.
The “brain-on-a-chip” contains hardware for machine learning and artificial neural synapses which help it analyze and quantify risks, adapt and respond to threats in real time, turning it into a “thinking” cybersecurity device.
CyberSwarm’s patented technology leverages biologically inspired hardware, specifically designed for AI, that enables devices to communicate with each other, identify what is going on in the network and react and adapt accordingly. The neuromorphic System-on-a-Chip generates a decentralized way of communication which is impossible to intercept. This facilitates completely private and tamper-proof communication between any two or more devices in a network.
Biologically inspired hardware AI
Neuromorphic engineering has given birth to chips whose architecture mimics that of the human brain. They are much more complex than traditional CPUs, allowing for more versatile and adaptable technologies. Responding to the needs of our time, neuromorphic chips process large quantities of information more quickly, with increased accuracy, solving two problems at once. It handles big data efficiently even offline and has an increased processing power with low energy consumption.
What this means for cybersecurity
By continously monitoring traffic inside the network, suspicious activity is immediately identified and stopped without the attacker knowing the alert has been triggered. The time required for internal investigations is reduced, and forensics are collected to be used in training the machine learning algorithms for future attacks. The hardware specialized for AI has the capability to process information without parsing it to a server, sense when a device is under attack and take the necessary steps to counter it.
The technology is protected through several US patents in the areas of neuromorphic computing and cybersecurity. More patents will be filed over the next period of time. We are committed to maintaining our technological advantage by continuously improving our product and innovating so that our customers can stay ahead of increasingly sophisticated security threats.
This is the first neuromorphic System on a Chip dedicated to cybersecurity, protecting from the hardware layer. It contains an array of artificial neural synapses which can mimic biological neuron function to create a “thinking” cybersecurity device which can act by itself in case of an attack. This offers local processing capabilities to the device, increasing its efficiency and accuracy while maintaining low power consumption. The SoCs communicate through a parallel communication channel which is impossible to detect for a third party, securing data exchanges.
Full control over data & communication
The multiple layers of security make communications between two or more devices in a network completely private and safe from 3rd party interference. Strong encryption makes the message impossible to decipher to anyone except the sender and the receiver.
Detect, respond and predict
Security teams can stay ahead of attackers by always knowing how far they are from critical systems and credentials. The predictive analytics feature enables them to anticipate and take deliberate action while there is still time, with little to no impact over the business flow.
Keep operations running smoothly
The technology has been designed as both an embedded technology and external add-on. It is based on a security model that is easy to implement, manage and scale. It eliminates the need for other security solutions (VPN, firewall, IDS, IPS), and customers can deploy their own software.
Today and tomorrow
The dedicated hardware can be easily integrated into existing IoT infrastructure, and also into future generations of devices without affecting any resources.
The underlying technology is highly versatile and can be applied to new products in many different areas of activity. Examples include: industrial equipment, medical equipment, consumer electronics, autonomous vehicles, guidance systems, etc.
The local processing capabilities are perfect for use in environments with limited resources, such as the Internet of Things or remote locations.
This eliminates the need for Cloud processing, creating a safer and more energy efficient alternative.
It is a great fit for companies looking to:
• securely transfer operations between Network Operations Centers – NOCs , workplace and datacenter
• safely deliver updates to equipment placed in locations that are difficult to reach
• safeguard their networks from intruders
• build their products with security in mind
Deliver Updates to Remote Equipment – make certain that critical equipment placed in locations that are difficult to reach receives updates that have not been tampered with.
Secure Networked Control Systems (NCS) – prevent attacks on critical infrastructure (power grid) by encrypting communications between sensors, controllers and actuators.
Industrial Equipment Security Add-On – it can act as a cybersecurity-dedicated add-on to machines that do not come with built-in security modules, such as industrial robots and drones.
Decentralized IT Networks – with the advent of 5G, the technology can be used to connect individual devices directly to the Internet, without having to use a router or modem.
Safe Cloud Computing – secure transmissions between operation centers using unbreakable encryption and gain control of how your data is stored in the cloud.
Wearables with Enhanced Privacy – encrypt the communication between medical equipment or other types of wearables that use the Internet to transmit sensitive data to servers.
San Mateo, California