3 ms·

I've read Kurzweil's books. Reflecting on them over time, I think he's kind of full of it. It's not that the material was un-interesting but rather that the c

I've read Kurzweil's books. Reflecting on them over time, I think he's kind of full of it. It's not that the material was un-interesting but rather that the claims he was making were a bit wild and overly-optimistic. I see the same kinds of flaws in the reasoning of many people of his generation, where they compile this list of truisms and try to get you from A->B but really the fundamental assumptions from A aren't that solid. Kurzweil wants to live forever. He wants to become a cybernetic being. That was his undoing in those books--his desire to outlast death forced him to make wildly optimistic predictions. The dangers facing us are not AI but simply mundane databases. Everything is being tracked these days. Compliance is being shoved down our throats. You are being watched on the Internet and in daily life. "But I have nothing to hide." Yes, you do. You just don't realize it. You should read up on how people were treated in East Germany and the Soviet Union because that is your future. Technology gives zero shits about humans.

spitfire 9y ago
I call these the Bhopal 2.0 problems. Bhopal was a disaster where 32tons of toxic MIC gas was released into the atmosphere and killed 3,787 injuring half a million people. My premise for Bhopal 2.0 is we automate processes so much and have no accounting for waste and damage that we simply focus on the top number. Ask an ML to optimize for an output and it will, happily, All the while harming people around the process - automatically. Right now Bhopol 2.0 processes are all focused on information systems - abstractions, so it's much harder to point and say "look, this is dangerous". People do very, very, very badly with abstractions. In some way this already happened with the recent elections. But that was purposely aiming the machines at us.
mr_spothawk 9y ago
> forced him to make wildly optimistic predictions. maybe I'm wrong, but I thought those predictions have fared pretty well. edit to add: optimism, or not... this came quickly (ahem, china) [1] > 2019, Public places and workplaces are ubiquitously monitored to prevent violence and all actions are recorded permanently. Personal privacy is a major political issue, and some people protect themselves with unbreakable computer codes. from his 1999 book, via wikipedia 1 - https://en.wikipedia.org/wiki/Predictions_made_by_Ray_Kurzweil#2019 https://en.wikipedia.org/wiki/Predictions_made_by_Ray_Kurzwe...

spitfire 9y ago

I call these the Bhopal 2.0 problems. Bhopal was a disaster where 32tons of toxic MIC gas was released into the atmosphere and killed 3,787 injuring half a million people. My premise for Bhopal 2.0 is we automate processes so much and have no accounting for waste and damage that we simply focus on the top number. Ask an ML to optimize for an output and it will, happily, All the while harming people around the process - automatically. Right now Bhopol 2.0 processes are all focused on information systems - abstractions, so it's much harder to point and say "look, this is dangerous". People do very, very, very badly with abstractions. In some way this already happened with the recent elections. But that was purposely aiming the machines at us.

mr_spothawk 9y ago

> forced him to make wildly optimistic predictions. maybe I'm wrong, but I thought those predictions have fared pretty well. edit to add: optimism, or not... this came quickly (ahem, china) [1] > 2019, Public places and workplaces are ubiquitously monitored to prevent violence and all actions are recorded permanently. Personal privacy is a major political issue, and some people protect themselves with unbreakable computer codes. from his 1999 book, via wikipedia 1 - https://en.wikipedia.org/wiki/Predictions_made_by_Ray_Kurzweil#2019 https://en.wikipedia.org/wiki/Predictions_made_by_Ray_Kurzwe...