Official statements are really laughable, especially the ones about safety. The system should be designed inversely: taking over the driver when in (serious) trouble. I even know the commercial name for it: HAD (human augmented driving). This approach could even safe lives instead of costing it.
You learn by your mistakes. If a system is not out there trying to constantly push the envelope regarding its capabilities, what’s the point of even running it?
Answer: hubris.
I winder which automated car companies have institutionalized hubris that would favor the perception of perfection over ongoing learning.. google/waymo? Uber? Hmm
StrangeDays said: Oh nos! “Apple is behind!” Where have we heard that one before? Oh, almost every single time. It’s like you guys have no memory of previous Apple launches and the chorus of identical comments.
Not something you want to rush.
No doubt... though I think this time they will all be failures, so I'd rather Apple be behind, and not wasting their money at all.
Official statements are really laughable, especially the ones about safety. The system should be designed inversely: taking over the driver when in (serious) trouble. I even know the commercial name for it: HAD (human augmented driving). This approach could even safe lives instead of costing it.
I'm not sure I want it completely taking over (if that's what you meant). Maybe you didn't follow a recent airline crash and the cause? (Or, did the MSM even report it correctly?)
But, I somewhat agree in terms of certain 'AI' assists to human driving. Just like *eventually* anti-lock brakes and especially stability controls are helping save lives, I think other assistive tech could be a huge benefit. For example, if we could detect a moose or deer off the road side and warn us, that would be nice. Or, if object detection gets good enough, automatic breaking under certain conditions would be nice too. Or, detecting if the driver is falling asleep, impaired, etc.
Safety is of course number one. I don't doubt Apple's stance there. But knowing Apple's interest in AR and machine learning, and AI, wouldn't it be safe to assume that the driver takeovers are not only for safety, as described in the report, but also for machine learning purposes? For teaching, and training the system? I can only assume that the self driving system is aware of the driver's actions, is recording it, analysing it, and ultimately learning from it.
Comments
The system should be designed inversely: taking over the driver when in (serious) trouble.
I even know the commercial name for it: HAD (human augmented driving).
This approach could even safe lives instead of costing it.
Answer: hubris.
I winder which automated car companies have institutionalized hubris that would favor the perception of perfection over ongoing learning.. google/waymo? Uber? Hmm
I'm not sure I want it completely taking over (if that's what you meant). Maybe you didn't follow a recent airline crash and the cause? (Or, did the MSM even report it correctly?)
But, I somewhat agree in terms of certain 'AI' assists to human driving. Just like *eventually* anti-lock brakes and especially stability controls are helping save lives, I think other assistive tech could be a huge benefit. For example, if we could detect a moose or deer off the road side and warn us, that would be nice. Or, if object detection gets good enough, automatic breaking under certain conditions would be nice too. Or, detecting if the driver is falling asleep, impaired, etc.