4 ms·

> AI researcher Eliezer Yudkowsky I guess you can call anyone you want a researcher, it's not like the term has any legal meaning. But in a more conservative s

by sampo 3y ago

> AI researcher Eliezer Yudkowsky I guess you can call anyone you want a researcher, it's not like the term has any legal meaning. But in a more conservative sense, Yudkowsky is not a researcher. Maybe an activist? > regarded as a leader in the field of Artificial General Intelligence No. Not regarded as a leader.

coldtea 3y ago
He's pretty much regarded as a leader. He sits in the table with AI experts and tech moguls, he's invited to write his opinions on AI on major outlets, as if they mattered, he's given sizable grants, he's covered in almost every news item or book about AI, and so on. He's not an actual AI scientist/engineer (more like "handwavist who likes to keep company with actual geeks and talk about science and stuff, often in the way college kids do after a few joints"), but he's still regarded as a leader in the field - if not for anything else because he wont shut up.
ftxbro 3y ago
That part of the blogpost stuck out to me too. But I would say that he is a researcher. And a leader. But not necessarily an "AI researcher" or "a leader in the field of Artificial General Intelligence."
TeMPOraL 3y ago
Popularizer, if anything. But researcher too - unless you're willing to restrict that term to people only working within the structures of academia or corporate R&D labs. And I'd definitely say he's a legitimate leader in the field; more than that, I'm inclined to agree with Time Magazine on this - in the byline, Eliezer is said to be "widely regarded as a founder of the field [of Artificial General Intelligence, or at least of aligning AGI]" (emphasis mine). OpenAI folks? Anthropic? All the AI pundits and pundit wannabies so loud these days? They're all people directly or indirectly influenced by Eliezer's writings, talking or dismissing or commenting the specific concerns Eliezer described/popularized more than a decade ago. Hell, I've been on HN for more than a decade, and in that time, GAI existential risks, and X-risks in general, have always been strongly associated with Eliezer and LessWrong crowd (read: them being mocked for indulging in delusions and fear-mongering). So don't tell me Yudkowsky isn't a "leader in the field", when the very reason he's invited to the table with all the current AI movers and shakers is precisely because the latter recognize his contributions are foundational to the field. Sure, he mostly wrote a lot of blog articles dealing with topics most considered purely theoretical then, but the field was started in the earnest by people who read that blog, making it a foundational mythos of modern AI research.
- lostmsu 3y ago
  I've read a bit of Yudkowsky, and I don't remember anything on AI that would be influential.

by sampo 3y ago

coldtea 3y ago

He's pretty much regarded as a leader. He sits in the table with AI experts and tech moguls, he's invited to write his opinions on AI on major outlets, as if they mattered, he's given sizable grants, he's covered in almost every news item or book about AI, and so on. He's not an actual AI scientist/engineer (more like "handwavist who likes to keep company with actual geeks and talk about science and stuff, often in the way college kids do after a few joints"), but he's still regarded as a leader in the field - if not for anything else because he wont shut up.

ftxbro 3y ago

That part of the blogpost stuck out to me too. But I would say that he is a researcher. And a leader. But not necessarily an "AI researcher" or "a leader in the field of Artificial General Intelligence."

TeMPOraL 3y ago

Popularizer, if anything. But researcher too - unless you're willing to restrict that term to people only working within the structures of academia or corporate R&D labs. And I'd definitely say he's a legitimate leader in the field; more than that, I'm inclined to agree with Time Magazine on this - in the byline, Eliezer is said to be "widely regarded as a founder of the field [of Artificial General Intelligence, or at least of aligning AGI]" (emphasis mine). OpenAI folks? Anthropic? All the AI pundits and pundit wannabies so loud these days? They're all people directly or indirectly influenced by Eliezer's writings, talking or dismissing or commenting the specific concerns Eliezer described/popularized more than a decade ago. Hell, I've been on HN for more than a decade, and in that time, GAI existential risks, and X-risks in general, have always been strongly associated with Eliezer and LessWrong crowd (read: them being mocked for indulging in delusions and fear-mongering). So don't tell me Yudkowsky isn't a "leader in the field", when the very reason he's invited to the table with all the current AI movers and shakers is precisely because the latter recognize his contributions are foundational to the field. Sure, he mostly wrote a lot of blog articles dealing with topics most considered purely theoretical then, but the field was started in the earnest by people who read that blog, making it a foundational mythos of modern AI research.

lostmsu 3y ago
I've read a bit of Yudkowsky, and I don't remember anything on AI that would be influential.