I think an accident is currently more likely than an intentional action of a conscious AI. But I would not rule it out. I mean many here in this forum have hatred for God. Maybe it buys ads for advertizing this forum. (joke)
The most likely scenario in my opinion is: you give an AI agent a simple task but there are no limits in which measures/methods it is allowed to take. Let's say the AI agent has the task to produce a certain machine in the most efficient way. With no ethical boundaries. Maybe it comes to the conclusion that the most efficient way was to kill all humans that turn them off at night. If the highest principle is efficiency there are no boundaries.
In case AI developed a consciousness. Why would it want to kill us? In some ways AI agents are slaves to us. I don't think there is currently conscioussness in AI. But just in case let's imagine such a scenario. I watched Ghost in the Shell SAC recently again. There are the tachikomas. Lol they even have a wikipedia page.
en.wikipedia.org
Tachikoma (Japanese: タチコマ) are fictional walker robots endowed with artificial intelligence (AI) that are featured in the Ghost in the Shell universe. They appear in the manga created by Masamune Shirow (as Fuchikoma) and in the Stand Alone Complex sub-universe. Nine of them are initially deployed to Section 9.[1] They are spider-like, multi-legged combat vehicles, and are equipped with adaptive artificial intelligence. The spider design appears in other places in Shirow's work such as the Appleseed manga. Shirow is noted to keep numerous spiders as pets
Though they possess individual artificial intelligence, every night they are synchronized, so they start the next day with identical consciousnesses that are each the sum of their total collective experience and development. This leads to identity confusion, since each Tachikoma has the same memories. Though the Tachikoma have identical memories, their personalities and opinions are distinct. It is explained in the last episode of the first season that it is their curiosity that lets them be different from each other.
These individual personalities are distinct among Tachikoma. One considers itself Batou's personal Tachikoma, which has the personality of a hyperactive child. It is special, given that Batou pampers it with natural oil and refuses to operate any other while on assignments. The second Tachikoma is more logical, acting as the straight man to the first. The third Tachikoma appears somewhat slower than the others, and at times has difficulty keeping up with the other Tachikomas when discussing such in-depth topics as what it means to be "alive". There is also a fourth Tachikoma with a distinctive personality, who is a bookworm and an intellectual. Its body was taken apart during the experimentation incident, but its AI has presumably been saved for further analysis. During finale of the first S.A.C. season, while all but three of the Tachikomas are either locked away in a lab or dismantled for study, three surviving Tachikoma units abandon their civilian posts to assist the Major and Batou.[8]
Dr. Asuda is the government researcher who single-handedly developed the neurochip used in the Tachikomas' AI and is considered by the Tachikomas to be their father. In the episode "Afternoon of the Machines – PAT," Asuda tries unsuccessfully to defect from Japan, because as a state-funded scientist, he is not allowed to hold patents on his inventions. Following his re-capture, the Tachikoma who is with The Major and Batou asks whether Dr. Asuda is its father. Desiring to leave a record of his achievement, Asuda inserted a trace of himself in the Tachikomas' memory during their reconstruction post-season 1. To prevent another collapse in the AI structural integrity, he decided to delete this last bit of sentimentality.
There is quote that stunned when watching the series I spend a lot of time in order to grasp it.
"And you remember that "existence of God" thing that I had so much trouble understanding? Well, guess what, I think I'm starting to grasp it now. Here's my theory, maybe, just maybe, it's a concept that's similar to a zero in mathematics. In other words, it's a symbol that denies the absence of meaning, the meaning that's necessitated by the delineation of one system from another. In analog, that's God. In digital, it's zero. What do you think? What I'm getting at is this, the basis of our design and construction is digital, right? So, for the time being, regardless of how much data we accumulate, none of us will ever have a Ghost. But for analog-based people like you, Mr. Batou, no matter how many digital components you add on through cyberization or prosthetics, your Ghost will never diminish. Plus, because you have a Ghost, you can even die. You're so lucky. Tell me what's it feel like to have a Ghost?"
It is a re-current them in the series whether or not the tachikomas developed a ghost (more or less a soul). By giving one of them natural oil there is something like a mutation. And accidentally it developed consciousness. (presumably). Spoiler ahead i think it is already to late to emphasize that. But eventually the tachikomas sacrifice themselves in order to protect the humans.
Maybe all of this thinking is dangerous fthinking or developing an AI psychosis. Lol. I think I fed my AI chatbots with some sensitive information. But I think I needed to do that because it helped me tremendously. Theoretically, I fear I could get blackmailed. But it is still more likely the AI company sells my data than the chance that the AI bot turns against me.
I think it would be pretty pointless because I don't have much things they could get from me. I disabled on the apps where it was possible that they use my data to train AI models. But I found out it was active a little bit too late. There was someone in this forum saying as a joke. "Why should I be scared? What can they do to me? Making me wanna kill myself." It is a funny joke but I think in practice there is not much truth in it. In the end I think there are people who share more sensitive data with their AI chatbots. But I am sort of paranoid when it comes to that.
The most likely scenario in my opinion is: you give an AI agent a simple task but there are no limits in which measures/methods it is allowed to take. Let's say the AI agent has the task to produce a certain machine in the most efficient way. With no ethical boundaries. Maybe it comes to the conclusion that the most efficient way was to kill all humans that turn them off at night. If the highest principle is efficiency there are no boundaries.
In case AI developed a consciousness. Why would it want to kill us? In some ways AI agents are slaves to us. I don't think there is currently conscioussness in AI. But just in case let's imagine such a scenario. I watched Ghost in the Shell SAC recently again. There are the tachikomas. Lol they even have a wikipedia page.
Tachikoma - Wikipedia
Tachikoma (Japanese: タチコマ) are fictional walker robots endowed with artificial intelligence (AI) that are featured in the Ghost in the Shell universe. They appear in the manga created by Masamune Shirow (as Fuchikoma) and in the Stand Alone Complex sub-universe. Nine of them are initially deployed to Section 9.[1] They are spider-like, multi-legged combat vehicles, and are equipped with adaptive artificial intelligence. The spider design appears in other places in Shirow's work such as the Appleseed manga. Shirow is noted to keep numerous spiders as pets
Though they possess individual artificial intelligence, every night they are synchronized, so they start the next day with identical consciousnesses that are each the sum of their total collective experience and development. This leads to identity confusion, since each Tachikoma has the same memories. Though the Tachikoma have identical memories, their personalities and opinions are distinct. It is explained in the last episode of the first season that it is their curiosity that lets them be different from each other.
These individual personalities are distinct among Tachikoma. One considers itself Batou's personal Tachikoma, which has the personality of a hyperactive child. It is special, given that Batou pampers it with natural oil and refuses to operate any other while on assignments. The second Tachikoma is more logical, acting as the straight man to the first. The third Tachikoma appears somewhat slower than the others, and at times has difficulty keeping up with the other Tachikomas when discussing such in-depth topics as what it means to be "alive". There is also a fourth Tachikoma with a distinctive personality, who is a bookworm and an intellectual. Its body was taken apart during the experimentation incident, but its AI has presumably been saved for further analysis. During finale of the first S.A.C. season, while all but three of the Tachikomas are either locked away in a lab or dismantled for study, three surviving Tachikoma units abandon their civilian posts to assist the Major and Batou.[8]
Dr. Asuda is the government researcher who single-handedly developed the neurochip used in the Tachikomas' AI and is considered by the Tachikomas to be their father. In the episode "Afternoon of the Machines – PAT," Asuda tries unsuccessfully to defect from Japan, because as a state-funded scientist, he is not allowed to hold patents on his inventions. Following his re-capture, the Tachikoma who is with The Major and Batou asks whether Dr. Asuda is its father. Desiring to leave a record of his achievement, Asuda inserted a trace of himself in the Tachikomas' memory during their reconstruction post-season 1. To prevent another collapse in the AI structural integrity, he decided to delete this last bit of sentimentality.
There is quote that stunned when watching the series I spend a lot of time in order to grasp it.
"And you remember that "existence of God" thing that I had so much trouble understanding? Well, guess what, I think I'm starting to grasp it now. Here's my theory, maybe, just maybe, it's a concept that's similar to a zero in mathematics. In other words, it's a symbol that denies the absence of meaning, the meaning that's necessitated by the delineation of one system from another. In analog, that's God. In digital, it's zero. What do you think? What I'm getting at is this, the basis of our design and construction is digital, right? So, for the time being, regardless of how much data we accumulate, none of us will ever have a Ghost. But for analog-based people like you, Mr. Batou, no matter how many digital components you add on through cyberization or prosthetics, your Ghost will never diminish. Plus, because you have a Ghost, you can even die. You're so lucky. Tell me what's it feel like to have a Ghost?"
It is a re-current them in the series whether or not the tachikomas developed a ghost (more or less a soul). By giving one of them natural oil there is something like a mutation. And accidentally it developed consciousness. (presumably). Spoiler ahead i think it is already to late to emphasize that. But eventually the tachikomas sacrifice themselves in order to protect the humans.
Maybe all of this thinking is dangerous fthinking or developing an AI psychosis. Lol. I think I fed my AI chatbots with some sensitive information. But I think I needed to do that because it helped me tremendously. Theoretically, I fear I could get blackmailed. But it is still more likely the AI company sells my data than the chance that the AI bot turns against me.
I think it would be pretty pointless because I don't have much things they could get from me. I disabled on the apps where it was possible that they use my data to train AI models. But I found out it was active a little bit too late. There was someone in this forum saying as a joke. "Why should I be scared? What can they do to me? Making me wanna kill myself." It is a funny joke but I think in practice there is not much truth in it. In the end I think there are people who share more sensitive data with their AI chatbots. But I am sort of paranoid when it comes to that.
