According to the study, in an accident where a human driver and artificial intelligence have dual control, the public is more inclined to blame the person.
The research team behind the study warned the attitudes could impact how juries apportion blame in any future fatal traffic accident cases involving autonomous and semi-autonomous cars.
The study, published in the Nature Human Behaviour journal, was carried out by experts from around the world, including from the University of Exeter’s Business School, the Massachusetts Institute of Technology (MIT), the University of British Columbia, Max-Planck Institute, Toulouse School of Economics, and Harvard University.
They asked members of the public to consider hypothetical cases in which a pedestrian was killed by a car operated under shared control between a human and autonomous technology.
It found that when either the human driver or the AI controls alone made an error, that driver was blamed more.
However, when both the human driver and machine made an error, the blame attributed to the AI was less.
Dr Edmond Awad, from the University of Exeter’s Business School, said: “It seems like if we leave it to the general public, they may unintentionally go soft on AV manufacturers to improve their safety standards.”
The researchers warned that current attitudes suggest a public under-reaction to any possible AI issues in driverless cars, with a greater focus instead on human accountability.
They argue that future regulation of such technology could fail to ensure the safety of such vehicles unless attitudes change.
A number of crashes involving autonomous and semi-autonomous vehicles have been reported in recent years, including the first widely publicised fatal incident in Florida in May 2016 when a Tesla driver using its semi-autonomous Autopilot feature hit the trailer of a large lorry.
An investigation by the US National Transportation Safety Board said the feature had played a role in the crash, prompting Tesla to roll out an update that disabled it if it detected the driver was not touching the wheel.