Even at 97% extinction risk, philosopher says AI could be worth it

Even at 97% extinction risk, philosopher says AI could be worth it

“Superintelligence” author Nick Bostrom calculates that people alive today could come out ahead even if the risk is very high

PHOENIX (AZFamily) — Philosopher Nick Bostrom has spent years warning that superintelligent artificial intelligence could threaten humanity’s survival. Now he argues that, at least from the perspective of people alive today, building it could be worth taking an extraordinary risk.

Bostrom, author of the influential 2014 book “Superintelligence: Paths, Dangers, Strategies,” made the case on Generation AI while discussing his new working paper.

The paper asks a deliberately narrow question: What path would maximize life expectancy for people who are alive now?

“The paper is sometimes misunderstood,” Bostrom said.

He said life expectancy is only one factor that should shape decisions about AI.

Bostrom said the risks of superintelligence are considerable. Some people at AI labs have floated odds of catastrophe above 10%, he said, and the real figure “might even be higher.”

But the status quo carries its own death toll. About 170,000 people die every day from disease, aging and other causes, Bostr...