Self-Learning AI learns your business
from the ground up to stop cyber disruption
While conventional AI techniques rely on pre-labelled data sets and prior knowledge, Self-Learning AI learns ‘on the job’ from real-world data and constantly adapts.
Developed by mathematicians
Darktrace’s founding mathematicians developed the AI algorithms that are now known as ‘Self-Learning AI’ in Cambridge in 2013. Under the leadership of Jack Stockdale OBE, the group used an interdisciplinary approach that brought together novel ideas and advances in machine learning, engineering, mathematics and computing infrastructure.
The technology saw early success when it was deployed at a major power station in the UK, where it helped identify highly unusual and threatening activity within the company that had gone unnoticed by traditional security controls.
Today, this fundamental technology powers the entire range of Darktrace security solutions, used across 6,800 organizations worldwide.
Core principles of Self-Learning AI
Bringing our Self-Learning AI to your data
In contrast to other AI approaches, which require data to be cleansed, labelled and moved to a centralized repository, Darktrace brings the AI to your data, wherever it lives.
Whether it’s in the cloud and on email systems, across Operational Technologies or traditional networks and infrastructure, Self-Learning AI can be applied to any of these systems, without requiring data migration. It learns from scratch, and constantly evolves its understanding as the data environment changes.
Self-Learning AI and cyber security
Thrives in complexity
Self-Learning AI thrives in complexity – the more users, devices and environments that the AI can analyze, the deeper its understanding becomes. It is constantly learning from new activity and data and builds greater cyber resilience over time.
Finds the unfindable
In cyber security, it’s impossible to second-guess tomorrow’s attackers. Self-Learning AI identifies patterns that no one thought to look for, enabling it to uncover novel attacker behaviors that would otherwise go unnoticed.
Highly accurate
Self-Learning AI is highly accurate in distinguishing between genuinely threatening behaviors, and unusual but benign activity. In fact, it’s so precise that it is capable of reacting to a cyber-threat without human intervention, while avoiding any disruption to the business.
At the heart of the Continuous Cyber AI Loop
Self-Learning AI powers a security solution ecosystem that allows organizations to prevent, detect, respond and recover from the full range of cyber-threats. Given its deep and interconnected understanding of the enterprise, Darktrace is uniquely positioned to deliver a virtuous cycle in which each capability is continuously optimized, with each AI brain feeding into the next.
Powering the world’s first Autonomous Response
Identifying where you have a real problem is critical. But today, detection is only half the battle. A cyber-attack can strike at any moment, and human teams cannot always react fast enough.
Self-Learning AI enables Autonomous Response, a technology that takes targeted action to interrupt cyber-threats in seconds.