Introduction

There is great promise for the use of Artificial Intelligence (AI) on our roads for assisted, augmented, adapted, automated and autonomous driving. AI can play a significant role to reduce road deaths and injuries (SDG 3.6) while also ensuring safe, affordable, accessible and sustainable transport systems (SDG 11.2).

However, to realise AI's potential to reduce the 1.3 million annual road deaths, it is important to define a minimal acceptable performance threshold that ensures AI systems continually perform at the level of a competent and careful human driver and never engages in reckless, dangerous or careless driving.

The widespread, socially acceptable, deployment of AI systems on our roads is dependent upon technology achieving public trust. During the AI for Good Global Summit 2019, the #AIforGood community called upon the ITU and the UNECE to establish an open forum and global dialogue between public, industry, regulators and standards bodies to establish a definition of this minimum performance threshold of AI systems on our roads. Ensuring that these AI ‘Drivers’ remain aware, willing and able to avoid collisions at all times.

Responding to this critical issue, the ITU and ADA have formed a new​ joint initiative to organise the first #AIforGood road safety event hosted at ITU Telecom in Budapest on Tuesday, 10 September. The objective of this ITU/ADA Workshop will be to present the state of the art in autonomous driving solutions and work together to define the “ADA Turing Test” for AI on our roads.​

Inputs from the ITU/ADA impact sessions and interactive afternoon workshop will be used to generate the first roadmap for international collaboration on creating top-down performance standards for AI on our roads.

"The Turing Test for Autonomous Driving - A Global Performance Standard for AI on our Roads"

Contact: tsbevents@itu.int