Free to use Taster Platform is live
CLICK HERE!🚀🚀🚀

We strive to build an environment where research intersects with industry, applying deep research to the societal concepts of fairness. With the growth of automated systems making decisions on the likes of insurance underwriting, medical resource allocation and hiring practices the time is now to ensure these systems are robust and fair.

Our Team

We are Fairgen.

Samuel Cohen and Nathan Cavaglione first met as class mates at Imperial College London studying Maths and Electrical Engineering respectively. Fast forward several years and they co-founded Fairgen, combining Sam's research experience with Nathan's time spent engineering product in a leading Israeli start-up. It was at this point they enlisted the support of Michael Cohen and Benny Schnaider, experienced entrepreneurs with proven track records for scaling organisations. Collectively they form Fairgen, a team of four tackling biased algorithms at source; the underlying data.

We've since built our team to 11 and are on a mission to bring our research backed approach to tackling bias in AI into the world.

Our mission.

Much of the focus on artificial intelligence has been driven by achieving performance improvements and in turn productivity gains at scale. However this poses a threat to individuals and the unique qualities they present, automated systems often come short when faced with diversity.

The time is now to ensure the systems being implemented to the world around us uphold if not further our human and societal values. We believe the introduction of artificial intelligence can further the progress we have made towards building a fair society.

Algorithmic Fairness

Automated systems should ensure those engaging them are treated fairly regardless of their profile.

Data Centric Solution

We believe that building a pre-processing, data centric solution will allow for the best results and seamless integration. It's our mission to build use case agnostic de-biasing technology.

Explainability

Decisions should be justified, many algorithms have to differentiate by nature, e.g Insurance Underwriting. However the pricing policy or output decision should be understood.