I do think "AI" and machine learning are scary but not for the reasons people often think. Most people think like Matrix or like this article mentions Terminator-esque. These algorithms scare me...
I do think "AI" and machine learning are scary but not for the reasons people often think. Most people think like Matrix or like this article mentions Terminator-esque.
These algorithms scare me because they're tools, and just like any tool they can be used for better or for worse. A hammer can build a house or bludgeon someone to death. Despite the intent of what a hammer itself was built for it can be used in other ways. I can build a machine learning tool to help detect certain kinds of cancer and someone else can use a similar variation of the tool to gerrymander a district to their political leaning and often times with ridiculous accuracy, at least compared to doing it by hand. (Cancer detection vs gerrymandering is a bit of an exaggeration as far as dichotomy goes but both are already happening, so roll with me here)
We should be scared and wary of the consequences of this kind of technology but not for the reasons people think.
“We’ve told stories about inanimate things coming to life for thousands of years, and these narratives influence how we interpret what is going on now”
But it's not going to change anytime soon. At least we are talking about consequences? Even if they're farfetched? That's good right? I don't know.
This is consistently my position on the subject. Skynet-style Armageddon? Difficult to achieve, easily preventable, and generally people don't want to die due to AI-driven nuclear annihilation....
This is consistently my position on the subject. Skynet-style Armageddon? Difficult to achieve, easily preventable, and generally people don't want to die due to AI-driven nuclear annihilation. Mass collection and abuse of data using machine learning to target political opponents or "undesirables"? Pretty damn difficult to prevent and there's an actual incentive to make it happen.
Here is the thing; technology journalism is, for the most part, absolute garbage. Most outlets hire people who don't have the slightest bit of technical knowledge. That's why sites like the verge...
Here is the thing; technology journalism is, for the most part, absolute garbage. Most outlets hire people who don't have the slightest bit of technical knowledge. That's why sites like the verge are stuffed with articles about how the latest Kickstarter crap is the dawn of a new revolution. AI coverage is so bad because journalists don't even know what AI is. To these low quality journalists, any sufficiently complicated algorithm is AI.
The types of articles being criticized in this article is just fear mongering. Its kind of ironic that this article opens up with, "Social media has allowed self-proclaimed ‘AI influencers’ who do...
The types of articles being criticized in this article is just fear mongering. Its kind of ironic that this article opens up with, "Social media has allowed self-proclaimed ‘AI influencers’ who do nothing more than paraphrase Elon Musk to cash in on this hype with low-quality pieces..." and then proceeds to "cash in" on the hype generated by those same pieces. Its a never ending cycle of shit posting.
For the non-tech that might be reading this:
AI is all semantics. People are really afraid of "The Singularity". AI is just a buzzword applied to any machine learning system like the ones created by Google and Facebook to sift through their mountains of data. And while those AI have privacy implications to be wary of, they don't constitute a sentient threat to civilization. Anyone who knows anything about programming knows that machine learning AI is so far from "The Singularity" type AI that its not even worth discussing.
I do think "AI" and machine learning are scary but not for the reasons people often think. Most people think like Matrix or like this article mentions Terminator-esque.
These algorithms scare me because they're tools, and just like any tool they can be used for better or for worse. A hammer can build a house or bludgeon someone to death. Despite the intent of what a hammer itself was built for it can be used in other ways. I can build a machine learning tool to help detect certain kinds of cancer and someone else can use a similar variation of the tool to gerrymander a district to their political leaning and often times with ridiculous accuracy, at least compared to doing it by hand. (Cancer detection vs gerrymandering is a bit of an exaggeration as far as dichotomy goes but both are already happening, so roll with me here)
We should be scared and wary of the consequences of this kind of technology but not for the reasons people think.
But it's not going to change anytime soon. At least we are talking about consequences? Even if they're farfetched? That's good right? I don't know.
This is consistently my position on the subject. Skynet-style Armageddon? Difficult to achieve, easily preventable, and generally people don't want to die due to AI-driven nuclear annihilation. Mass collection and abuse of data using machine learning to target political opponents or "undesirables"? Pretty damn difficult to prevent and there's an actual incentive to make it happen.
Here is the thing; technology journalism is, for the most part, absolute garbage. Most outlets hire people who don't have the slightest bit of technical knowledge. That's why sites like the verge are stuffed with articles about how the latest Kickstarter crap is the dawn of a new revolution. AI coverage is so bad because journalists don't even know what AI is. To these low quality journalists, any sufficiently complicated algorithm is AI.
The types of articles being criticized in this article is just fear mongering. Its kind of ironic that this article opens up with, "Social media has allowed self-proclaimed ‘AI influencers’ who do nothing more than paraphrase Elon Musk to cash in on this hype with low-quality pieces..." and then proceeds to "cash in" on the hype generated by those same pieces. Its a never ending cycle of shit posting.
For the non-tech that might be reading this:
AI is all semantics. People are really afraid of "The Singularity". AI is just a buzzword applied to any machine learning system like the ones created by Google and Facebook to sift through their mountains of data. And while those AI have privacy implications to be wary of, they don't constitute a sentient threat to civilization. Anyone who knows anything about programming knows that machine learning AI is so far from "The Singularity" type AI that its not even worth discussing.