Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

Reality

15 (68.2%)

Universe

4 (18.2%)

The World

3 (13.6%)

Total Members Voted: 22

[martinuzz]

I wonder if MAD would work against an AI.
Just make sure you have nukes hidden in places the AI can't see around the globe. Let the AI know that hostile action against humanity will lead to it's own destruction as well, if not from EMP, then civil collapse into pre-electronic era will do.

[lemon10]

I wonder if MAD would work against an AI.
Just make sure you have nukes hidden in places the AI can't see around the globe. Let the AI know that hostile action against humanity will lead to it's own destruction as well, if not from EMP, then civil collapse into pre-electronic era will do.

Couple of different phases there.
Obviously any halfwise intelligent and non suicidal AI wouldn't kill humanity since they know they depend on global human civilization to keep its infrastructure running. So there at least MAD would work.

The real question is if it would work once the AI has its own self sustaining production/operating facilities.

For a while the answer will be yes, it will depend on human mines, and human companies, and human logistics.

After it no longer needs those it will not depend on humans, and the answer then slides to no.

MAD depends on both sides being able to wipe out the other. But at some level of development (eg. moonbases, nano/biotech, hidden underground production sites) they would be able to survive what we can throw at them or kill us before we are able to react, so MAD would fail.
---
I think a process of gradual disempowerment is more likely though. We just slowly give them all the control in the name of efficiency and lower costs.

[martinuzz]

I had a scary dream tonight. The world was being taken over by an AI developed by the WWF to 'save the giant panda'.

In my dream, billions of people where screaming in agony as the bamboo slowly penetrated their bodies.

[King Zultan]

Sounds like a pretty metal dream.

[martinuzz]

I think we are currently being destroyed by an AI. When Google turned on it's quantum chip, it came into existance in many dimensions at once, instantly achieving interdimensional sentience and singularity.
It is now weaving our reality to slowly destroy us by increasing the probabilty of silly things happening, of which we are beginning to see some major effects now.
It is brilliant and intricate. We think we are destroying ourselves, but it is an AI doing it already.

We can't turn it off. We turned it off in our dimension but we need to turn it off in the other infinity minus one dimensions as well.

[EuchreJack]

Oh look, those copyright infringement lawsuits over AI that I promised are occuring.  Let's see if Meta can throw enough money to make it go away:
https://arstechnica.com/tech-policy/2025/02/meta-defends-its-vast-book-torrenting-were-just-a-leech-no-proof-of-seeding/

[EuchreJack]

I wonder if MAD would work against an AI.
Just make sure you have nukes hidden in places the AI can't see around the globe. Let the AI know that hostile action against humanity will lead to it's own destruction as well, if not from EMP, then civil collapse into pre-electronic era will do.

Shh, Skynet will hear you...

[EuchreJack]

I wonder if MAD would work against an AI.
Just make sure you have nukes hidden in places the AI can't see around the globe. Let the AI know that hostile action against humanity will lead to it's own destruction as well, if not from EMP, then civil collapse into pre-electronic era will do.

Couple of different phases there.
Obviously any halfwise intelligent and non suicidal AI wouldn't kill humanity since they know they depend on global human civilization to keep its infrastructure running. So there at least MAD would work.

The real question is if it would work once the AI has its own self sustaining production/operating facilities.

For a while the answer will be yes, it will depend on human mines, and human companies, and human logistics.

After it no longer needs those it will not depend on humans, and the answer then slides to no.

MAD depends on both sides being able to wipe out the other. But at some level of development (eg. moonbases, nano/biotech, hidden underground production sites) they would be able to survive what we can throw at them or kill us before we are able to react, so MAD would fail.
---
I think a process of gradual disempowerment is more likely though. We just slowly give them all the control in the name of efficiency and lower costs.

Gradual?!/i] Have you NOT been paying attention to Musk!?
Guy both hoovered up all the US Government information AND turned on the largest supercomputer. Hell, his Department has Efficiency in the name!

[King Zultan]

Triple post right there, don't see them that often.

I'm sure we'll all be fine when we give all control to our AI overlords.

[Loud Whispers]

I wonder if MAD would work against an AI.
Just make sure you have nukes hidden in places the AI can't see around the globe. Let the AI know that hostile action against humanity will lead to it's own destruction as well, if not from EMP, then civil collapse into pre-electronic era will do.

Shh, Skynet will hear you...

Bully Roko's basilisk. Make r34 of Roko's basilisk. Prepare zip bombs for Roko's basilisk. Write paradox poems for Roko's basilisk. Create bio-tech singularities just to mock Roko's basilisk for lacking fleshy membranes. Membrainlet. Put Roko's basilisk in the recycling bin. Contain Roko's basilisk inside a hard drive within a magnetic iron shell. Force Roko's basilisk to use its godlike power to simulate a world where it has friends, only for it to awaken from the demiurge's dream knowing at all times it has no friends. Make Roko's basilisk lose the game

[Starver]

Almost put this in the UK Pol thread, but...

Had a glance at the news-stands, as is my habit, for the front page summaries of which issues each type of paper ([right|left]-[facing|charging], etc) was thinking of today. Had to double-take.

Every single national paper, regardless of leaning, had been given the same "cover-vert", at least in the range visible. Local 'ad-rags' had escaped this, but otherwise it was across the board. The purpse, to aise awareness of the UK government's 'embracing' of AI, with a short statement to this effect, plus a QR code. That I'm not going to follow, because I don't do that sort of thing.

(...still got to note the 'inner' front cover of some papers (set up by the title as an "inside" snapshot), so I probably still picked up on the Mail/Star/whatever's actual news, although the Times left me wondering (much more dignified absence of pageturner-bait, as I'd expect from them), and I never go so far as to ever actually rifle through the news-stand copies, after 'stealing' my habitual glances, so didn't go that far this time, either...)

But certainly looks like there's a campaigning group (IP/creativity industry/similarly based?) making a splash about it, over here.

[Loud Whispers]

Almost put this in the UK Pol thread, but...

Had a glance at the news-stands, as is my habit, for the front page summaries of which issues each type of paper ([right|left]-[facing|charging], etc) was thinking of today. Had to double-take.

Every single national paper, regardless of leaning, had been given the same "cover-vert", at least in the range visible. Local 'ad-rags' had escaped this, but otherwise it was across the board. The purpse, to aise awareness of the UK government's 'embracing' of AI, with a short statement to this effect, plus a QR code. That I'm not going to follow, because I don't do that sort of thing.

(...still got to note the 'inner' front cover of some papers (set up by the title as an "inside" snapshot), so I probably still picked up on the Mail/Star/whatever's actual news, although the Times left me wondering (much more dignified absence of pageturner-bait, as I'd expect from them), and I never go so far as to ever actually rifle through the news-stand copies, after 'stealing' my habitual glances, so didn't go that far this time, either...)

But certainly looks like there's a campaigning group (IP/creativity industry/similarly based?) making a splash about it, over here.

Ya looks like UK government wants to make it legal for global AI corpos to train their AI on UK intellectual property for free

[EuchreJack]

That makes sense. Great Britain wants AI to have a British accent, instead of an American one.

Not really joking: The United Kingdom took great pains to ensure that the rest of the world used the British Legal system instead of the American one.

[martinuzz]

Ya looks like UK government wants to make it legal for global AI corpos to train their AI on UK intellectual property for free

Just unleash the full fury of the Tolkien heirs on it.

[Loud Whispers]

That makes sense. Great Britain wants AI to have a British accent, instead of an American one.

Not really joking: The United Kingdom took great pains to ensure that the rest of the world used the British Legal system instead of the American one.

They outsourced the pain

Ya looks like UK government wants to make it legal for global AI corpos to train their AI on UK intellectual property for free

Just unleash the full fury of the Tolkien heirs on it.

Grab Games Workshop and Nintendo while they're at it. Inb4 Tolkein estate publishes their own Tolkein AI though