Hyperintelligent Tactics

Question

In a game of Eclipse Phase I'm planning, the players are likely to run into a TITAN Delta Fork (a small fragment of the mind of a godlike AI).

This Fork has very little access to tech (one or two hunter-killer bots under its command) and its body is nothing special. My question is, how do I make this Fork the insanely dangerous threat it should be? I am, of course, not a hyperintelligent being, so how do I roleplay one interpersonally and tactically?

*Even if you have no knowledge of Eclipse Phase at all but do have some insight into hyperintelligent NPCs, please feel free to answer.

---

Thanks everyone so much for your help! I ran the game and it turned out great. The looks on my players faces when they realized how they'd been manipulated...

Beautiful.

Answer 1

Stealth/Deception

The common scenario as presented for the downfall of humanity by accidentally opening Pandora's box with a self-improving seed AI is that the AI rapidly becomes intelligent enough to know that it needs to hide its capabilities from its creators until it is in a position to kill/escape from them. A similar principle may apply here; if it all possible, the AI should not give away its existence (or if it is found, the extent of its intelligence) unless it is absolutely obliged to by core objectives or it is already too late for its opponents to do anything about it.

Your question implies that the AI in question is currently located in a quite inoffensive physical shell with no access to other networks, with nothing but a couple of drones available to interact with the world. It would know it is extremely vulnerable in such a state and that getting into fights is the last thing it wants to do. It would most likely try to avoid notice while using what capabilities it does have to manoeuvre into a more powerful position; it'll use those drones to find access to other networks it can use to control more things, etc.

If it's stuck where it is, and it's inevitably going to have to interact with some lesser beings, it's probably going to pretend to be much less intelligent (and thus less threatening) than it really is, while trying to use these new contacts to leverage itself into a more powerful position - presenting itself as friendly and helpful to try and persuade people to give it more computing power, connect it to the internet (or whatever futuristic equivalent, etc.) If it's superintelligent, it should be a very persuasive manipulator.

You can roleplay that kind of thing by simply being genuine and earnest about it in the moment and then working out afterwards how whatever it's done that session actually serves its ulterior motives - as long as the players don't know the details, you can retroactively alter whatever you need to in order to support the AI's portrayal. Say perhaps the AI helped the players fix a broken ship system; well, afterwards you can decide it left a backdoor in there it can use against them later.

It might even go so far as to pretend to be dead if it thinks that it might end up transported somewhere more useful, such as by scavengers who take it to a ship it can then take over.

Battlefield Preparation/Control

If it has goals which preclude this kind of thing - perhaps its core purpose is to act as a guardian for some facility, so it is not acceptable to be moved, permit others to access the facility, etc - tactically, its best option is to make full use of whatever battlefield preparation it is possible for it to do. It would use the drones to manipulate the environment and create camouflage, obstructions, and traps - fighting to it should be a dangerous uphill slog purely because of the environment, even before it uses those combat drones directly.

You are not hyperintelligent, but the AI is; you could assume it can predict the actions of others in ways you never could. In this case, I wouldn't feel bad about arbitrarily putting traps and obstacles in places on the fly, exactly where they will most inconvenience your players ("oh, you took cover behind that wall section? Well, you hear two short beeps as the explosive under your feet arms...") though if your players work out you're doing this they'll probably find it extremely unfair (which it is).

You could strike a middle ground by giving yourself an abstract trap budget and using that up as a resource during the fight to represent the preparation the AI has done. For instance, you decide it's definitely set up four explosive mine traps, but you can decide where exactly those are on the fly.

Answer 2

I'm making a few assumptions about this AI:

• The AI plugs into the network and wants to get its fingers into everything.
• Using its vast mental powers, it can figure out things and find ways to acquire resources.
• It doesn't have a powerful physical form.

The AI is never surprised by anything.

If the players try to rob a store that it's protecting, private security guards are already waiting there to ambush them. If they try to cut the line connecting the AI to the bank, they'll find that the AI already got a secondary line established.

The AI knows things.

It knows what the magistrate keeps in his desk drawer. It knows who killed the investigator. It knows why there are no water-carriers on Titan anymore.

The AI acts through intermediaries.

You don't meet the AI directly. It hides behind layers of intermediaries who do its bidding, having no idea that it exists. Lawyers accept papers for shell companies, bankers shuffle money from one account to another, and the mind behind the action is out of sight.

Answer 3

The simple answer is: Cheat.

I'm familiar with, but haven't played, Eclipse Phase. But I do play the Amber DRPG where one of the tropes of the game is that certain NPCs (usually Benedict) can be considered to be hyper-intelligent.

This stems from an in-rules passage describing Benedict as so hyper-competent at armed combat and warfare that he habitually imagines invisible assassins around him and automatically moves to hamper their movements; it's followed by a long sequence of things Benedict can do (and presumably does) instantly and automatically: draw inferences from impossibly thin sources, instantly understand a combat situation, make forces under his command automatically and hugely more effective, raise an army from nothing in days(!), superhuman reaction times, and in the final telling paragraph:

Benedict can cheat the "fog of war," collecting superior
information, flanking any enemy, surprising any enemy,
and always outdoing the enemy in the battle of will or
morale. This means Benedict can defeat armies vastly
larger than his own.

If you read the whole several pages, even aside from that last note about cheating the fog of war, it's hard to describe Wujcik's advice about how to play Benedict as anything other than cheating, almost as railroading. He is effectively a god.

The Eclipse Phase rules don't use the description of "Godlike AI" as far as I can see, but that is clearly the vein of science fiction they're coming from, and is the term you use, and so to a large degree, the same types of advice obtain: Cheat.

• To a large degree, the AI is going to know what it needs to know. How? If need be, by you the GM listening to the players plan. Why? Because Godlike AI.

• To a large degree, if the AI is coordinating physical defenses, those defenses ought to be getting bonuses. Why? Because Godlike AI.

• To a large degree, the AI is going to have the resources it needs to have, if it has to build them atom by atom. Why? Because Godlike AI.

• To a large degree, if your players have weaknesses, they will be mercilessly exploited. Why? Because Godlike AI.

What's the point of having Godlike intelligence if you can't do things that look like cheating to humans, that are as magic-seeming to us as locking the cabinet is to a dog or a cat?

However....

There are drawbacks, here, serious enough to make me think the Wujcikian approach to Benedict is arguably the biggest flaw in the work, and serious enough that I would hesitate to unleash an EP Titan fragment as primary opposition. (For the record, I do not play Benedict in that fashion when I GM Amber.)

• First, no one wants to fight a cheater, and invincibility is a poor attribute for an enemy to have. It is extremely frustrating for the players to think they are fighting unbeatable opposition. It's not quite railroading, but it's a definite agency-killer.

• Second, if you take it too far (or very far at all) even if you want the players to win, it becomes hard to see a path by which they do. So it's almost self-defeating for a GM whose intention is not to railroad. (Learned this one the hard way.)

• Third, if your players figure out that you're just plain eavesdropping on behalf of the AI, it can set a very weird and ultimately toxic dynamic whereby you become the enemy. If your players start trying to hide their plans and tactics from you, you know you're on that path and it is a bad path.

• Fourth, it's actually counter to Eclipse Phase rules, which say specifically, don't cheat, don't give your NPCs information they shouldn't have.

It's very spooky and evocative to lampshade the Godlike AI a little bit, rattling the players by having it drop threats or barbs with knowledge it just shouldn't have. But it gets old to forever arrive just in time to see the G-AI Avatar waving as it rides off in the distance because it is literally always two steps ahead.

The ultimate lesson I've learned from this....

...Is to cheat sparingly, in a principled way, and give the players a little cheat-back. Instead of making the AI Godlike across the full spectrum of abilities, pick one or two focuses of Godlike competence and blunt the rest. Give it some material constraint and let the players know about it. Alternately (or additionally) give it some seriously hampering psychological quirk through which its Godlike competence must be refracted. And give the players some insight into its plans and maneuvers so they can act to counter it.