I'm surprised this hasn't gained more traction! Anyway, it's interesting how the narratives align with the incentives here: The "incumbents" (Anthropic, OpenAI et al) want regulation that protects them from competition, so they predict doom if development is allowed to continue unchecked. Nvidia wants to continue selling hardware to whoever is leading the AI race, so Huang is more optimistic.
"Oil drill manufacturer says there is 0% chance global warming is real"
"Gun retailer says guns kill people not guns"
Why are we listening to Jensen Huang's opinion on this? The bias is palpable.
Of course AI has dangerous aspects (the gun metaphor is perhaps apt here), but asking the man who stands to gain the most from unregulated AI growth is not the way.
I give it non-zero chance. But not because AI becomes sapient and does something. But because either economy gest sufficiently screwed or some idiot massively missuses AI...
Then again economy is already screwed. And idiots could always do something bad. So nothing new really.
My p(doom in my natural lifetime) risk is about 10%, which is mostly the chances we keep ignoring all the warnings as they get more capable and thus their errors more damaging.
I'm surprised this hasn't gained more traction! Anyway, it's interesting how the narratives align with the incentives here: The "incumbents" (Anthropic, OpenAI et al) want regulation that protects them from competition, so they predict doom if development is allowed to continue unchecked. Nvidia wants to continue selling hardware to whoever is leading the AI race, so Huang is more optimistic.
"Oil drill manufacturer says there is 0% chance global warming is real" "Gun retailer says guns kill people not guns"
Why are we listening to Jensen Huang's opinion on this? The bias is palpable.
Of course AI has dangerous aspects (the gun metaphor is perhaps apt here), but asking the man who stands to gain the most from unregulated AI growth is not the way.
The bias of Anthropic and OpenAI and anyone associated with longtermism, transhumanism and effective altruism is massively higher then his.
I give it non-zero chance. But not because AI becomes sapient and does something. But because either economy gest sufficiently screwed or some idiot massively missuses AI...
Then again economy is already screwed. And idiots could always do something bad. So nothing new really.
I'm an optimist by the standards of people who have ever written "p(doom)", I'd put p(doom before 2030) at 2-3% or so.
On that scale, mostly political risk, e.g. this kind of thing going hot: https://www.yahoo.com/news/us/articles/fake-ai-intelligence-...
My p(doom in my natural lifetime) risk is about 10%, which is mostly the chances we keep ignoring all the warnings as they get more capable and thus their errors more damaging.
hallucination
But, on the other hand, chance that Trump, Thiel and Vance in cooperation with Russia, AfD and the rest of gang will destroy it anyway is pretty high.
[dead]