3 ms·

One major problem of automotive transport is society's denial of the risks involved. Mandating showing autonomous interventions while hundreds die every day un

One major problem of automotive transport is society's denial of the risks involved. Mandating showing autonomous interventions while hundreds die every day un-viewed in traditional cars will make everyone less safe.

jmpman 7y ago
For all I know, my politicians have been incentivized to allow these autonomous cars on my streets. (For politicians, all it takes is the ability to tout - It’ll create jobs, you’ll be able to get on TV, looking like an innovator) Why should I believe the government that they’re safe? (The most recent Uber self driving death felt like a coverup being encouraged by the politicians as the original unredacted video wasn’t shared for a long time) How can the public become confident that the autonomous cars are safer than human drivers? Certainly there’s an incentive for the autonomous vehicle manufacturers to lie or distort the trust? What if these autonomous vehicles are just safer than a drunk driver blowing 0.2 BAC, but not as safe as a drunk driver blowing 0.08? If presented with those metrics, would you be so dismissive of their risks? Without human drivers intervening, I expect self driving cars outside of boring highway miles to be currently equivalent to around a 0.3 BAC. Yes, I understand the point that self driving are an incredible potential life saving tool, and when perfected, thousands of lives will be saved. But if they’re released too early (know of any company with a financial incentive to release a product too early?) they are going to have many victims - possibly someone I know. And yes, the safety of autonomous cars driving on bumper to bumper LA traffic is going to be used to justify their safety on single lane highways during a Michigan snowstorm. Remember that when self driving companies are citing their “better than human drivers” claim, the humans they’re comparing against is “all humans”. Who causes the human accidents? The drunk, distracted, inexperienced or very old. Eliminating those drivers from the statistics, incorporate non-boring highway miles from the statistics, and you’ll quickly find that human drivers are likely tens of thousands of times safer than autonomous cars. Isn’t that a story the public should understand? But, if you add on automatic red light/stop sign detection and auto braking to augment a human driver, even the drunk, distracted, inexperienced and very old will become 1000x better than today, and that’s going to save lives. Don’t become blinded by the starry future. It’s coming, but before we get there, it’s more likely to be an the fire of an autonomous car that mistook a fire engine as a drivable highway.
simion314 7y ago
But a new problem is people thinking computer are always right so they follow the GPS navigation blindly over a cliff or they don't pay attention when a poorly named driver assist is engaged. As long as self driving companies release PR materials (and the fanboys as well) there should be fine with everyone to see the other side too. Btw the claim that autopilot is better then an average human was not proven yet, all the existing statistics are very misleading comparing some new cars versus all vehicles in all conditions. Or the other statistic from Tesla is under doubt due the fact when asked for the actual data to be made public Tesla refused (let me know if there was any progress in this case)

jmpman 7y ago

For all I know, my politicians have been incentivized to allow these autonomous cars on my streets. (For politicians, all it takes is the ability to tout - It’ll create jobs, you’ll be able to get on TV, looking like an innovator) Why should I believe the government that they’re safe? (The most recent Uber self driving death felt like a coverup being encouraged by the politicians as the original unredacted video wasn’t shared for a long time) How can the public become confident that the autonomous cars are safer than human drivers? Certainly there’s an incentive for the autonomous vehicle manufacturers to lie or distort the trust? What if these autonomous vehicles are just safer than a drunk driver blowing 0.2 BAC, but not as safe as a drunk driver blowing 0.08? If presented with those metrics, would you be so dismissive of their risks? Without human drivers intervening, I expect self driving cars outside of boring highway miles to be currently equivalent to around a 0.3 BAC. Yes, I understand the point that self driving are an incredible potential life saving tool, and when perfected, thousands of lives will be saved. But if they’re released too early (know of any company with a financial incentive to release a product too early?) they are going to have many victims - possibly someone I know. And yes, the safety of autonomous cars driving on bumper to bumper LA traffic is going to be used to justify their safety on single lane highways during a Michigan snowstorm. Remember that when self driving companies are citing their “better than human drivers” claim, the humans they’re comparing against is “all humans”. Who causes the human accidents? The drunk, distracted, inexperienced or very old. Eliminating those drivers from the statistics, incorporate non-boring highway miles from the statistics, and you’ll quickly find that human drivers are likely tens of thousands of times safer than autonomous cars. Isn’t that a story the public should understand? But, if you add on automatic red light/stop sign detection and auto braking to augment a human driver, even the drunk, distracted, inexperienced and very old will become 1000x better than today, and that’s going to save lives. Don’t become blinded by the starry future. It’s coming, but before we get there, it’s more likely to be an the fire of an autonomous car that mistook a fire engine as a drivable highway.

simion314 7y ago

But a new problem is people thinking computer are always right so they follow the GPS navigation blindly over a cliff or they don't pay attention when a poorly named driver assist is engaged. As long as self driving companies release PR materials (and the fanboys as well) there should be fine with everyone to see the other side too. Btw the claim that autopilot is better then an average human was not proven yet, all the existing statistics are very misleading comparing some new cars versus all vehicles in all conditions. Or the other statistic from Tesla is under doubt due the fact when asked for the actual data to be made public Tesla refused (let me know if there was any progress in this case)