Why the hell do people think that POD CASTING is the shizzle? Every time I turn around, someone has a new pod cast. Most of them are trash, if we're honest here. I listen to only two or three regularly. All these celebs be like "people need to hear my opinions and they need to hear me shoot the shit with my fellow celebs."
I like Adam Carolla, because he's a funny dude. I listen to the Lanterne Rouge because it's the best cycling related podcast. Also, have I said that I F***ing hate Lance Armstrong? I do. That guy is a POS, a cheat and a fraud. He can die in a fire, for all I care.
Most podcasts seem to be 'celebrities' or "journalists" without the clout or connections to get paid to journalism. There are a few good ones (one more example of Sturgeon's Law). All of them move too slow for me. Sometimes if there is a transcript I'll read that.
But I don't care about the opinions of celebrities, failed politicians, and most failed journalists.
The AI story makes total sense. My wife and I were out and about, she was driving, and when she went to parallel park the parking assist suddenly activated and the car said, "Listen toots, you better just leave this one to me, capeesh?"
AI is racist and sexist because it can accurately make use of basic statistics?
You train an algorithm in the US where black men commit crimes at a higher rate than any other combinations of race/sex, and you're surprised when they show criminals as black men at a higher rate than any other combination of race/sex? If you trained the AI in different country, where whites commit crimes a higher rate than blacks, I'm confident that the AI would pick white men most often as the face of crime (I'm fully confident that only in Themyscira do women commit more crime than men on average) and nobody would say mention racism.
And just how many men identify as "homemaker?" Change it to "homebuilder" and I bet AI would be picking the face of high testerone men for that label.
re: racist and sexist robots. I don't know whether the reporter is really this dumb (just like the AI though, I'm making a good guess) or it wasn't explained to him well. "visually classifies objects, is built by scraping billions of images and text captions from the internet" means the AI was essentially 'trained" to associate some images to some words depending on the frequency with which it had encountered the 'connection' within the "billions" it looked at.
It's neither racist nor sexist, it is showing/telling you what billions of images and words and bazillions of connections have shown to it.
Just like humans, if you force it to guess, it will go with the highest likelihood IN IT'S EXPERIENCE.
If I were forced to bet my life on which person out of, say, eight was a criminal and told that the images were selected totally randomly except for the criminal, I'd damn sure go with the probabilities.
Anyone who says they wouldn't is a d*mn liar or has serious mental problems.
The guy attacked Zeldin with a cat-shaped self-defense tool made for women and designed to scratch out a mugger's eyes with the pointy ears. It's dangerous, but it makes this attacker look hilariously pathetic.
Why the hell do people think that POD CASTING is the shizzle? Every time I turn around, someone has a new pod cast. Most of them are trash, if we're honest here. I listen to only two or three regularly. All these celebs be like "people need to hear my opinions and they need to hear me shoot the shit with my fellow celebs."
I like Adam Carolla, because he's a funny dude. I listen to the Lanterne Rouge because it's the best cycling related podcast. Also, have I said that I F***ing hate Lance Armstrong? I do. That guy is a POS, a cheat and a fraud. He can die in a fire, for all I care.
I’ve wondered that about celebrity podcast myself. Why the hell do I care about what some “celebrity” thinks? I’d rather listen to
Someone with some expertise on the subject. I get tired of being lectured to by morons. Same on Armstrong
Most podcasts seem to be 'celebrities' or "journalists" without the clout or connections to get paid to journalism. There are a few good ones (one more example of Sturgeon's Law). All of them move too slow for me. Sometimes if there is a transcript I'll read that.
But I don't care about the opinions of celebrities, failed politicians, and most failed journalists.
If Chris Cuomo wants to be there for his brother, that's great, but it means he's not a journalist.
The AI story makes total sense. My wife and I were out and about, she was driving, and when she went to parallel park the parking assist suddenly activated and the car said, "Listen toots, you better just leave this one to me, capeesh?"
AI is racist and sexist because it can accurately make use of basic statistics?
You train an algorithm in the US where black men commit crimes at a higher rate than any other combinations of race/sex, and you're surprised when they show criminals as black men at a higher rate than any other combination of race/sex? If you trained the AI in different country, where whites commit crimes a higher rate than blacks, I'm confident that the AI would pick white men most often as the face of crime (I'm fully confident that only in Themyscira do women commit more crime than men on average) and nobody would say mention racism.
And just how many men identify as "homemaker?" Change it to "homebuilder" and I bet AI would be picking the face of high testerone men for that label.
Yay for the Wonder Woman reference!
re: racist and sexist robots. I don't know whether the reporter is really this dumb (just like the AI though, I'm making a good guess) or it wasn't explained to him well. "visually classifies objects, is built by scraping billions of images and text captions from the internet" means the AI was essentially 'trained" to associate some images to some words depending on the frequency with which it had encountered the 'connection' within the "billions" it looked at.
It's neither racist nor sexist, it is showing/telling you what billions of images and words and bazillions of connections have shown to it.
Just like humans, if you force it to guess, it will go with the highest likelihood IN IT'S EXPERIENCE.
If I were forced to bet my life on which person out of, say, eight was a criminal and told that the images were selected totally randomly except for the criminal, I'd damn sure go with the probabilities.
Anyone who says they wouldn't is a d*mn liar or has serious mental problems.
The guy attacked Zeldin with a cat-shaped self-defense tool made for women and designed to scratch out a mugger's eyes with the pointy ears. It's dangerous, but it makes this attacker look hilariously pathetic.