3 ms·
To play devil's advocate: Assuming self-driving cars are connected to each other, either directly or via the internet, if one car crashes, or even just makes a
by LeoPanthera 8y ago
To play devil's advocate:
Assuming self-driving cars are connected to each other, either directly or via the internet, if one car crashes, or even just makes a minor mistake, the knowledge of how to avoid that can be transmitted to every other car on the planet.
- mc32 8y agoJust as good information can be propagated, so can bad. So this had better be a robust algorithm and not one which can either be induced into something or simply learn “wrong”.
- valine 8y agoSeems unlikely that the knowledge of how to avoid making a mistake could be compiled without human analysis. The technology to do that autonomously doesn’t exist. It would almost require artificial general intelligence, and if we had that self driving would be trivial.
- bdamm 8y agoAutomatic feature selection is the area of research you’re looking for.
- londons_explore 8y agoThe "human analysis" could be completed within 30 seconds and beamed out to all cars in the area. No reason Tesla can't have operatives on duty 24/7 to assist and tweak/train the fleet of cars.
- natch 8y agoThis is very close to what is actually happening, though not in real time.
- craftinator 8y agoBased on what evidence?
- natch 8y agoBased on what Tesla presented in their autonomy day event (https://www.youtube.com/watch?v=Ucp0TTmvqOE https://www.youtube.com/watch?v=Ucp0TTmvqOE) but it's also what they've been saying publicly for a while now... in a nutshell, they ask the cars to send selected summary data back when interventions happen and in certain other cases.
- craftinator 8y agoAh thank you. I had only ever heard that it was a beta feature, something they planned to implement in the future.
- grey-area 8y agoTesla demonstrated a system doing exactly this at their autonomy event. The video is worth watching whatever you think Tesla.
- noelsusman 8y agoThis feels more dangerous, not less.