Our Product

Spherical Defence is a unique API security system that uses machine learning to understand the language of the application. Spherical Defence automatically builds a model of the application’s API, evolving dynamically so it can recognize legitimate data traffic and detect intrusions. This learned knowledge maintains application security as it adapts to changes in hackers’ behaviour, and can subvert hackers by generating spoofed responses that mimic the behaviour of the API.


In recent benchmarked trials on live application data, Spherical Defence achieved an intrusion detection score of 99.9998% on a dataset of 50 million logs, rendering it ten times more effective than the traditional web application firewalls.

Rapid Deployment

Our model is built using API access logs, which may be historic data, or real time API requests. Unlike WAFs, there is no need for the creation of rules or signatures

Easy Integration

Our technology fits within your existing infrastructure, be that on premise or a private cloud.  Our models are agnostic to your choice of infrastructure

Secure and Confidential

All your data stays within your network, and our model can be built and operated without requiring any third-party access to your data

Unattended Learning

We dynamically build any number of models to protect each of your applications, without the need for user intervention

Transparent Operation

Security is provided with little or no performance degradation.

Resilient

We provide fail-safe service against any single point of failure

How does it work?

Spherical Defence employs a multitude of sophisticated methods to provide efficient API security. See below:

Learning Traffic

We combine variational inference with natural language processing techniques to learn a probability distribution over the live traffic in your service. This distribution encompasses the structure and content of fields and values, understanding questions such as: What keys does your data contain? How entropic are the values for this field? Using this information we identify malicious traffic on an individual basis.

Learning Behaviors

Data habits evolve over time, and we take that into consideration. Using stochastic process analysis, a tool commonly used to describe Chaos Theory, we evaluate how the dialogue between your application and an end user is changing. This allows Spherical to create an understanding of your app's behaviour and protect against behaviour based attacks (e.g hackers stress testing your application).

Learning to Learn

There is no ‘one size fits all’ machine learning system that protects every API equally well. Each requires subtle differences in their parameters (eg, learning speed, size of model, depth of model) to suit the unique characteristics of an individual application. At Spherical, we apply genetic algorithms that learn the best parameters in an evolutionary context, dynamically ruling out bad parameter sets and building the best mix of parameters. Over time, our model becomes optimised for each application, and continues to learn long after being deployed on the client’s infrastructure.