News for hackers. Stuff that matters.
Re:Alexa suggests lethal challenge to child by cortesoft on Tuesday December 28, 2021 @12:35AM (#29710814)
I mean, regular old intelligence was used to create the original message that this artificial intelligence then repeated.

Also, if people are worried that a person would fall for this challenge, then clearly intelligence doesn't always catch this misinformation.

So since both regular intelligence and artificial intelligence can be fooled by this, I don't see how you can determine that artificial intelligence doesn't know what it is doing any more than regular intelligence also doesn't know what it is doing.

Re:Alexa suggests lethal challenge to child by pdonis on Tuesday December 28, 2021 @12:52AM (#29710900)
> if people are worried that a person would fall for this challenge

The person who asked Alexa the question was 10 years old. That's why people are worried.

Re:Alexa suggests lethal challenge to child by cortesoft on Tuesday December 28, 2021 @01:27AM (#29711077)
Sure, but a 10 year old has real intelligence... they have an idea of what is going on, no one is arguing they don't have the same kind of intelligence as adults, just that they have things they still need to learn.... so why couldn't you say the same of the AI that is learning how to give challenge recommendations? Just because it isn't perfect doesn't give evidence for or against artificial intelligence being a real thing.
Re:Alexa suggests lethal challenge to child by charcircuit on Tuesday December 28, 2021 @01:58AM (#29711225)
>then clearly intelligence doesn't always catch this misinformation.

It's not misinformation as it is an actual challenge, albeit a potentially harmful one.