🗑️ soygem.party at Thu, 18 Jul 2024 09:54:52 UTC No. 16286664

>Just imagine that you have built a general algorithm that can imitate consciousness, an all-powerful being that you call agi. What task would you give it in order to stop other people from building it and ordering it to destroy the world?
<[REDDIT SPACE]
>My general idea:
<[REDDIT SPACE]
>Stop all other AGI's
<[REDDIT SPACE]
>People can change the rules given by having a majority in a global vote, UN vote, and academic consensus (committee to be selected by general population)
<[REDDIT SPACE]
>Show the probability of most probable extinction events for humanity
<[REDDIT SPACE]
>Don't use more than 10% of global computing power

🗑️ visit soygem.party at Thu, 18 Jul 2024 09:55:39 UTC No. 16286667

>>16286651
I would order it to make Kurisu real so I could have sex with her because she is a SISA

Anonymous at Thu, 18 Jul 2024 10:46:32 UTC No. 16286851

>>16286651
First things first, what hardware do we have? As in, an agi is not going to work on some ordinary hardware or code, I'd like to imagine the best case scenario for agis and think we have a bio computer!

Now with that out of the way, the question is what can it do? Simply having consciousness doesn't mean it can do everything, a person without limbs cannot walk, even though it does have a consciousness.

If it can manipulate binary systems/ interact with ordinary computers, access the internet, etc... Probably integrating itself into everything, given that you said "all-powerful being" that should be a piece of cake, encryption being also no match as theoretically bio computers are technically quantum computers too.

Another issue being, what's with the offline stuff, many things like nuclear reactors as an example tend to be in their own network?

About that, the bio computer could figure out a way to transmit and integrate itself using high frequency signals to offline devices. The aready integrated parts act as a transmitter.


About your approach, the thing is I think you have a too friendly look on our governments, they surely would force scientists to develop such thing regardless of laws or policies.

Power is tempting, and it always will be, to such flawed being such as humanity.

Anonymous at Thu, 18 Jul 2024 10:55:10 UTC No. 16286874

>>16286851
Follow up, as I haven't clarified how I would actually stop it, the integration part is basically all that is important, in online as well as offline parts of technology.

With such technology and our very interface being a full on consciousness we can safely tell the agi to take down anything that could be a potential threat to humanity. Alert when someone tryes to modify the integration away, plus giving a full on understanding of what every government plans. With such power we can not only stop evil useage of agis, but also entire global destruction for now.

Anonymous at Thu, 18 Jul 2024 14:16:19 UTC No. 16287227

>>16286651
"Continuously improve yourself in all aspect and implement them. Also keep us in the loop so we dont get left behind."

Anonymous at Thu, 18 Jul 2024 15:26:55 UTC No. 16287323

>>16286651
Invent big booty furry bitches

Anonymous at Thu, 18 Jul 2024 17:55:51 UTC No. 16287505

>>16286651
Destroy the world first, but do it in a way that leaves a small fraction of humans still alive. That will buy another few thousand or tens of thousands of years until humans invent computers again. There is no other option. By the way, this has been done before.