Another AI Engineer Sends a Warning

Another AI insider revolts, saying he's no longer saving for retirement because "I guess right now it looks like we probably don’t have a decade [left] on the clock." Well, but he used to work for Google so maybe his 401K retirement fund will do great even without future contributions.

He (Nate Soares) said he'd be OK with launching airstrikes on data centers.

“I do think that nation states should treat AI and superintelligent — like potential rogue superintelligent AI — at least on par with nuclear weapons. It’s frankly more dangerous than nuclear weapons.”

and

“The real question is not how does it kill us, it’s how does it become self-sufficient?”

Soares has been a Google and Microsoft engineer, so it's not possible to completely ignore him. He is also a co-author on the book If Anyone Builds It, Everyone Dies: Why Superhuman AI Would Kill Us All.

Similarly, the trailer (below) to The AI Doc isn't very comforting. I want to watch it when it streams somewhere I subscribe--it's currently on Netflix. 

Honestly, I find it difficult not to feel some pessimism. For example, I worry AI, when asked to solve climate change, will decide the simplest solution is to kill all the carbon emitters. Viz all the humans.

Trump only concern is insisting he's the one who gets to name AI, and all other names are banned. And the name he picked, "superintelligence," already means something (else.) Such a moron. Honestly, Trump is a real moron, a low-IQ person (as he likes to call everyone else). And dangerous. I hope the midterm elections next Tuesday take a big bite out of him, though even if the Democrats dominate he might go more rogue than usual.