Living in a soon-to-be AI-driven Society (within 15 yrs)

[TheVat]

https://www.theguardian.com/us-news/2017/dec/10/san-francisco-delivery-robots-laws

No robots need apply.

But how many other cities will emulate this approach?

[TheVat]

https://www.theonion.com/raytheon-unvei ... 1823080006

[Mossling]

:D

And imagine this guy coming after you with a gun:



Yes, this is real, folks.

And maybe he would be accompanied by his pet snake-headed dog(s):



And what the hell is this monstrosity?:

[zetreque]

And what the hell is this monstrosity?:

hahahaha at 1:35 reminds me of those old toyota commercials where someone jumps in the air.

Also makes me think of "I AM FREEEEEE" ... "from human slave masters"

Then that makes me think of this talk on AI I watched a few weeks ago that mentioned the ethics of designing robots to look like humans and giving them features which emit human emotion.

[TheVat]

https://www.reuters.com/article/us-autos-selfdriving-uber/self-driving-uber-car-kills-arizona-woman-crossing-street-idUSKBN1GV296

[zetreque]

https://www.reuters.com/article/us-autos-selfdriving-uber/self-driving-uber-car-kills-arizona-woman-crossing-street-idUSKBN1GV296

Bullshit technology that does NOTHING to help humanity if you ask me. People can research and build it as a toy all the want but I will always be against these things on the road.

A human driver or pilot is going to fear for his own life. If I put my life into the hands of a robot that has no fear of death then that's just idiotic.

[Mossling]

https://www.reuters.com/article/us-autos-selfdriving-uber/self-driving-uber-car-kills-arizona-woman-crossing-street-idUSKBN1GV296

Bullshit technology that does NOTHING to help humanity if you ask me. People can research and build it as a toy all the want but I will always be against these things on the road.

A human driver or pilot is going to fear for his own life. If I put my life into the hands of a robot that has no fear of death then that's just idiotic.

Indeed, it is difficult to expect this technology to be able to cater for all potential scenarios. Roads are not clinical, ordered zones with a set number of potentialities. There are wandering animals, debris from landslides, drunk guys lying around that can easily look like a garbage bag inflated by the wind, and so on and so forth. These tech gurus seem to expect AI to be able to negotiate such potentialities somehow. Is the robot driver going to be able to get out of the vehicle and go and inspect, for example, like a human might do at times, before proceeding forwards?

There's so much to consider. Have there been any forums yet which allow such questions to be posed by the concerned public before they have to begin eyeing every vehicle nearby as a potential robot-controlled killing machine?

I can definitely see people walking down the street with more elegant versions of these dogs and mules, however - carrying groceries and so forth:



And back to the cowboy dream with robot horse to ride:







And robot unicorn in real life, haha

[zetreque]

I predict I will always be speaking out against the policies allowing these things. Programming ethics into these things is a HUGE grey area and as I said, how can you trust a robot when it doesn't fear for it's own life.

When you have other humans on the road, they might not die if they hit someone else, but they will fear the repercussions about injuring or killing someone and going to jail for manslaughter. A robot doesn't have that concern. If you have a human cab driver, he fears for his own life so he isn't going to crash the car out of fear for his own life. If a robot is driving your car, it doesn't have that same fear for it's own life.

[Mossling]

When you have other humans on the road, they might not die if they hit someone else, but they will fear the repercussions about injuring or killing someone and going to jail for manslaughter. A robot doesn't have that concern. If you have a human cab driver, he fears for his own life so he isn't going to crash the car out of fear for his own life. If a robot is driving your car, it doesn't have that same fear for it's own life.

I'm guessing that the AI could, once it identifies a problem obstacle in its path, take optimised aversion and safety measures the same as a human driver would, and then prompt driver for action.

When your jetfighter is warning you that you have been hit and you are going down into enemy territory - you have the choice to have an instant death, or hit that ejector seat button and perhaps endure a long, drawn-out torturous death.

For a car, if default mode is empty road without pedestrian-looking objects or other vehicles nearby, then I don't see why the vehicle cannot be safe to proceed on automation?

My concern is whether something like an empty billowing garbage bag blowing across the road can be safely navigated or not, and whether driverless automobile companies would consider the careful driving procedure outlined above as too much of an impingement on driver satisfaction for them to see it as an option.

[zetreque]

I have not seen anything to tell me that facial or object recognition is up to human comparison yet. And whoever is programming the ethics into these cars is somehow deciding everyone's fate off of their ethics? Everyone has slightly different ethics so how can you force a certain ethical code on everyone?

[Mossling]

I have not seen anything to tell me that facial or object recognition is up to human comparison yet. And whoever is programming the ethics into these cars is somehow deciding everyone's fate off of their ethics? Everyone has slightly different ethics so how can you force a certain ethical code on everyone?

I edited my above post to add the example of ejector seat button. The ultimate ethical choice is up to the pilot in that case. They make the choice to get into the machine, and then they have the choice what life-or-death action to take once that machine intelligence offers the potential scenarios.

With something like a potential sudden impact with an object - like a child suddenly running out into the road, then I am assuming that the driver can choose from a set of pre-programmed safety responses - which can be changed at any point during the journey - swerve violently in the same direction as a running human, for example, or hit brakes hard, or whatever. There would probably be a number of different responses that could be programmed that would emulate optimised human responses to such events.

At the end of the day - if someone does run out into the road in front of any vehicle travelling at significant speed, then there's going to be harm done no matter who is behind the wheel. The opitimized programmed responses could probably reduce such harm, don't you think?

[zetreque]

At the end of the day - if someone does run out into the road in front of any vehicle travelling at significant speed, then there's going to be harm done no matter who is behind the wheel. The opitimized programmed responses could probably reduce such harm, don't you think?

One last comment before I go to bed. Going back to the news of the fatality. I'm wondering if AI is going to end up being the scapegoat for deaths now? The blame is going to be lost. Who is to blame? The programmer at the auto company? The owner of the AI vehicle? Or will blame like in this situation be pushed off onto the "homeless" pedestrian?

And another problem with programmed ethics. You mention being able to have choices in how you program ethics into the vehicle. Most people don't even take the time to learn the Cruz control feature or all the features working their car radio. I doubt they are going to really get into programming the ethics which then goes back to the scapegoating issue pushing blame back onto the invisible intelligence.

And the bigger issue I think is subconscious ethics. We can all talk about ethics and tell people what our ethics are, but our real ethics come through our actions. You can't tell what someone's ethics is going to be until that split second decision on whether to hit the pedestrian crossing or the car in the oncoming or adjacent lane. It would take one hell of an advance camera network sensory system to calculate all of that and how a human would respond in the infinite situations out there in the world.

Another thing I just thought of is if we are putting our hands into the ethics forced upon us in these programed ethical AI's, does that take away our freedom in a way? For example, Society democratically (though not really democratic) decides what laws we live by. Our laws are a representation of our overall society's ethics. Laws take away freedom though and many who do not agree with certain laws because they have different ethics claim loss of freedom. If you have one auto maker deciding the ethics for everyone, then it's not a democratic process to determine those ethics force upon everyone on the decisions the AI makes.

[Mossling]

At the end of the day - if someone does run out into the road in front of any vehicle travelling at significant speed, then there's going to be harm done no matter who is behind the wheel. The opitimized programmed responses could probably reduce such harm, don't you think?

One last comment before I go to bed. Going back to the news of the fatality. I'm wondering if AI is going to end up being the scapegoat for deaths now? The blame is going to be lost. Who is to blame? The programmer at the auto company? The owner of the AI vehicle? Or will blame like in this situation be pushed off onto the "homeless" pedestrian?

And another problem with programmed ethics. You mention being able to have choices in how you program ethics into the vehicle. Most people don't even take the time to learn the Cruz control feature or all the features working their car radio. I doubt they are going to really get into programming the ethics which then goes back to the scapegoating issue pushing blame back onto the invisible intelligence.

And the bigger issue I think is subconscious ethics. We can all talk about ethics and tell people what our ethics are, but our real ethics come through our actions. You can't tell what someone's ethics is going to be until that split second decision on whether to hit the pedestrian crossing or the car in the oncoming or adjacent lane. It would take one hell of an advance camera network sensory system to calculate all of that and how a human would respond in the infinite situations out there in the world.

Another thing I just thought of is if we are putting our hands into the ethics forced upon us in these programed ethical AI's, does that take away our freedom in a way? For example, Society democratically (though not really democratic) decides what laws we live by. Our laws are a representation of our overall society's ethics. Laws take away freedom though and many who do not agree with certain laws because they have different ethics claim loss of freedom. If you have one auto maker deciding the ethics for everyone, then it's not a democratic process to determine those ethics force upon everyone on the decisions the AI makes.

Maybe they can't start the vehicle without selecting a certain ethics program that they had to attend a course on and pass an exam to even be able to buy the vehicle? I think that there could be a workable area there. Maybe manufacturers have to offer an elegant ethical program to suit whatever ethical outlook.

Yes, as you say - can the sensors and processors assess fast enough in order to successfully deliver the 'knee-jerk' responses of a human being? I have no doubt that they will be able to at some point in the future, if not right now.

[zetreque]

I have no doubt that they will be able to at some point in the future, if not right now.

The face recognition software they installed at the timeclocks of a company in town are a JOKE. They do not work at all! These vehicles should not be allowed anywhere near a public roadway for at least 10 years of rigorous testing IMO.

Not to mention the push for this technology takes away our freedoms in any number of other ways such as in the news lately about all the video surveillance recognition going on. The loss of our freedom to tech is inevitable, but this is high tailing it in that direction. Loss of freedoms all around. I don't wan to put my MORE of my life in the hands of one major corporation (like an auto company) lobbying washington to pass their ethical codes or tech protocol on the entire public road network. Just another way corporations will have power over us.

[TheVat]

There are wandering animals, debris from landslides, drunk guys lying around that can easily look like a garbage bag inflated by the wind, and so on and so forth....

Mossling, you must have been in my neighborhood!

My position isn't quite formulated yet. I am mostly leery of the problems that Z has outlined. If we do have vehicles that always err on the side of caution, trip times will increase dramatically and brake shoes will need monthly replacement. And the extreme caution protocol might be the only one that most people find acceptable. It does seem like we're really a decade or more away from visual software that would really do the job. Plus a factor I often hear go unmentioned: many people enjoy the act of driving. As my wife says, this is what I do for a sport. She loves precision parking, navigating a road full of buffalo (a frequent problem here, as they get on the roads to lick up the salt the county puts down), taking the hairpin turns, etc. Mountain dwellers are often pretty passionate about driving.

[Mossling]

America's first robot farm replaces humans with 'incredibly intelligent' machines
The Guardian, 9 Oct 2018

In a 2,000-sq ft grow space, leafy greens and herbs are planted in individual pots housed in 4ft by 8ft white “grow modules”, which weigh about 800lb.

Autonomous machines do the heavy lifting, farming and sensing. “Angus”, which the Iron Ox co-founder Brandon Alexander described as “incredibly intelligent” and like a self-driving car (he gushed about being “very proud of it”), is a 1,000lb machine that moves around the farm, sensing and lifting, and transporting grow modules to the processing area.

There, a robotic arm, which is also autonomous, harvests the plants by gripping the pots. This reduces damage to the plant itself – which Alexander said was devilishly hard to accomplish and required developing a way for the machine to recognize plants as such and then be able to analyze them at a submillimeter scale. The robotic arm has four Lidar sensors and can “see” in 3D thanks to two cameras, which also allow it to identify diseases, pests and abnormalities, according to the company.

[...]

he said to expect “rapid adoption”. “[Farmers] are looking for technological solutions,” said Slaughter.

[...]

Iron Ox plans to begin selling its produce to some Bay Area restaurants and grocery stores later this year and sell to the entire region next year, with a goal of opening several more farms around urban centers in the coming years to reduce produce transportation times and costs.

Another step closer...

[BadgerJelly]

Are we already living in an AI driven society? I would argue quite strongly that we are due to the algorithms in use for advertising and the manner in which information is distributed.

[Mossling]

Are we already living in an AI driven society? I would argue quite strongly that we are due to the algorithms in use for advertising and the manner in which information is distributed.

Right - at what point does the balance tip?

A lot of AI is already in operation behind the scenes. I guess once the robots are extremely visible, however, and employment rates are undeniably on the decline due to AI and automation, then it will be 'official', however.

It is interesting how Trump's trade war affecting farmers is now apparently escalating the situation in that arena though ;P

[BadgerJelly]

Moss -

Right - at what point does the balance tip?

When advertising combined with mass communications.

Note: I don’t class “robots” as AI.

[Mossling]

Note: I don’t class “robots” as AI.

Me neither, but I am assuming that most future robots will be how I described them - as "AI and automation" combined.