London, Aug 25 : Researchers have uncovered significant fairness issues related to age and skin colour in the detection systems of autonomous vehicles, revealing that children and individuals with darker skin are at more risk on the street, a new study has shown.

According to the study conducted by researchers at King's College in London, a fairness analysis of eight different AI-powered pedestrian detectors trained on "widely-used, real-world" datasets revealed that the programmes were significantly worse at detecting darker-skinned pedestrians than lighter-skinned pedestrians. 

They found through testing over 8,000 images through these pieces of software that detection accuracy for adults was 19.67 per cent higher compared to children, and there was a 7.52 per cent accuracy disparity between light-skinned and dark-skinned individuals.