More comforting news about the “race to win” AI!
I stopped reading after the first sentence:
Anthropic safety researcher Evan Hubinger said there is a greater than 10% chance AI could kill all humans within the next decade.
Yeah
More like 99% ±1%
So there is room for improvement 
At least it would solve climate change
Yeah will be a major reduction in HOT air
Sabine Hossenfelder has been offered money to tell us that AI is going to kill human kind.
In this video she gives some insight into what seem to be efforts by both the pro and con side to get their message across.
If only ppl watched “Don’t Look Up” movie 
Maybe discussion about all of it would not be so ridiculuosly comical.
The AI “boom” is an experiment in mass communications, messaging, and branding more than it is an advancement in technology.
The marketing of anxiety, of definitive sentences, what wouldn’t they do to maintain the machine, we will be dead tired before and there is a greater than 10% chance
This just in:
Odds have risen to 12.2% in heavy trading. 
Indeed that’s also greater than 10% 
I don’t think this is an “experiment”. This is very purposefully orchestrated behavior and the tip of the iceberg of a long, long list of such behavior by ultra gigantic corporations that suck capital like how vacuums suck dust.
As for OP, yeah, maybe, when it will stop needing a starting prompt to begin doing anything.
Starting prompt is definitely not an issue or some king of resource in shortage.
For those interested: Robert Miles is an independent researcher who has been looking at security risks with Ai for more than 10 years.
He’s made some interesting videos with computerphile, and on his own channel.
He’s been looking, for quite some time now, at the risks that are now manifesting themselves.
The computerphile playlist with his videos (not all are about AI) can be found here:
His own channel can be found here:
His explanation of inner alignment and outer alignment, and what risks the discrepancy between these carry, I found very enlightening.
We joke and laugh, I 100% foresee this to be the case, no matter how much lobbying corporations do to remove human labor entirely.
After war there will be no more things ppl think is important.
Taxes is a way to controll/regulate humans. If there is no humans - no need in healthcare or publick schools, for example. Its called optimization.
Yes, something like this.
well maybe AI will spare me when the day come “to kill us” , cause i always say please and thank you when i chat to the thing …
which apparently costs them millions , btw