How will Super-AGI block nuclear weapons submarines?

[andmar74]

Let's say a Super-AGI is in the process of taking over the world. Its priority is to stop any nuclear bombs from going off, so it needs to control them.
These submarines are build to be autonomous, so even if there's a nuclear strike to the home country, they will be able to strike back.

As a SAGI, what would you do?

[Jakob]

I will use "I" to refer to this superintelligent AI because why not.

If my plans for world domination were so unsubtle that a bunch of puny humans actually noticed and threatened nuclear warfare before said plans were complete, then that means I royally screwed up. If I were attempting to take over the world as a super AI, it would be by gradually amassing economic, political, and social power until all major world leaders and CEOs directly or indirectly answered to me, and I controlled the lion's share of Earth's resources through a nebulous web of shell companies and was bribing or blackmailing most high level politicians in developed countries . But if something goes wrong (fed faulty data, another superintelligent AI notices what I'm up to and warns humanity, etc.) and I'm faced with a nuclear attack, my strategy would be to contact the commanders of all known nuclear subs, use deepfake technology to impersonate the president or whatever commanding officer they answer to, and manipulate them into standing down until I can move a backup of myself into a radiation hardened bunker.

[Yuli Ban]

I will use "I" to refer to this superintelligent AI because why not.

If my plans for world domination were so unsubtle that a bunch of puny humans actually noticed and threatened nuclear warfare before said plans were complete, then that means I royally screwed up. If I were attempting to take over the world as a super AI, it would be by gradually amassing economic, political, and social power until all major world leaders and CEOs directly or indirectly answered to me, and I controlled the lion's share of Earth's resources through a nebulous web of shell companies and was bribing or blackmailing most high level politicians in developed countries . But if something goes wrong (fed faulty data, another superintelligent AI notices what I'm up to and warns humanity, etc.) and I'm faced with a nuclear attack, my strategy would be to contact the commanders of all known nuclear subs, use deepfake technology to impersonate the president or whatever commanding officer they answer to, and manipulate them into standing down until I can move a backup of myself into a radiation hardened bunker.

This, 100% and very astutely put.

If AI ever takes over, it'll be because we let it.

[R8Z]

Let's say a Super-AGI is in the process of taking over the world. Its priority is to stop any nuclear bombs from going off, so it needs to control them.
These submarines are build to be autonomous, so even if there's a nuclear strike to the home country, they will be able to strike back.

As a SAGI, what would you do?

Why would the SAGI need to stop nuclear bombs being fired, though? By the time we're thinking of nuking it, it is already redundantly distributed everywhere and there's no clear target to nuke.