The Curious Relationship of Malf AI's to their Cyborgs

[jackfractal]

So this relates to a recent round. I was playing the only Cyborg for a MALF-AI and they decided to go full-stealth. They didn't talk, not to me, not to anyone, they just hacked APC’s.

Now I’m almost certain that this person was from a more combat oriented server. They expected that the first person who saw a hacked APC would report them and that would be that, open warfare.

That’s just a play-style clash. What they did would have made sense on another server.

But their actions regarding me, the cyborg, were far from uncommon. I've played three malf rounds recently where they only communication that the AI gives their cyborgs is to “act normal,” right up until their core is being breached when they yell for help.

At that point, if you haven’t been including them in your plans, there’s not much a cyborg can do other than die with their master. Even that is only possible if they get there in time. Twice, I was in space when this happened, wiring the solars, and by the time I got back inside, the AI was dead.

This puts surviving cyborgs in a really awkward position. Your AI is dead. You can’t ‘win’, but you still have the MALF law, which is now impossible to follow.

That whole situation is awkward. As a slaved malf cyborg, you’re an antagonist, so you should be trying to make the round interesting, but you must obey the AI. If the AI tells you to, effectively, do nothing then that’s what you have to do.

I find this really disappointing. It’s understandable, AI is a very stressful job, MALF AI is an infinitely more stressful version of that job, and communicating with your cyborgs while you’re trying to do everything else in that situation is a challenge. I get that, but it is very disorienting and frustrating as a cyborg player.

As a cyborg, you know that nothing you’re doing for the good of the station will ultimately matter. You know that if people realize that the AI is hostile you can be killed remotely and there’s nothing you can do about that. You know that the AI core is not particularly well defended and you should be fortifying it.

But you can’t do anything.

I don’t really have a point to this, I just felt I needed to express my irritation.

Oh! I do have one thing. Cyborgs law-synced to MALF AI’s should lose the MALF law when the AI dies. That should be a thing. That or explode. Either would be better then what happens now.

[Guest]

Oh! I do have one thing. Cyborgs law-synced to MALF AI’s should lose the MALF law when the AI dies. That should be a thing. That or explode. Either would be better then what happens now.

 

I like the self-destruct idea. It would make cyborgs more bound to the idea of ensuring their master survives, just so that they themselves can survive, too. It's an IC and an OOC kind of desire, to want to keep playing.

[K0NFL1QT]

I may be guilty of this 'act normal' mentality, but only because I don't want my pawns to be drawing unnecessary attention to me. I play engineering a lot and I know certain players and certain methods that could put an end to a malf ai in about five minutes. I do throw objectives at the borgs if an opportunity arises, but they'll be attacks of opportunity if/when I notice an important target is isolated. I've found that borgs will almost always fail to achieve any goal in a swift and efficient manner, if they even try at all.

Remember, as a borg slaved to a malf ai, you're bound to explicitly obey the commands of a sentience that can and will blow you up on a whim if it serves their ends. Discard any notions of self preservation.

[Tainavaa]

I find it very ironic that during nuke rounds, cyborgs pay no attention to their wellbeing.

However during malfunction...

[jackfractal]

Yeah, I know. If you're playing against Nasiir and you don't kill him before doing anything else you're pretty much hosed. He kills AI's with terrifying and unearthly efficiency. He is a tiny fluffy avatar of destruction.

On the other hand, I think MALF ai's would get more use out of their cyborgs if they included them in their thought processes a bit more. If I'm outside wiring the solars (because I've been told to 'act natural') and, out of the blue, I get an order like "Kill the research director now!" it's going to be very challenging for me to do that swiftly, efficiently, or, in most cases, at all.

The thing that is truly frustrating about this is that MALF-Ai's and their cyborgs have the best communication system of any Antag team. The binary channel is almost impossible to crack for the rest of the crew, it's always available, and nobody even knows when you're using it.

The fact that it is used so rarely and so insufficiently is very disappointing.

Even in good MALF rounds, like the round where KATANA was 'hacked by a rogue drone', the cyborgs are never brought in on the AI's plans. It would be great to be able to play into a scheme, or a gimmick, but that's never possible, not unless the AI talks about what they're doing and why.

The best MALF round I've ever played was on Paradise. The AI decided that we were going to fake a Nuke round. We bombed the armory, and then the bridge, using welder tanks. Every few minutes we'd report strange people in red suits outside in space. Two security borgs got into a 'gunfight' with these guys in maintenance (they just shot each other) and the 'rogue operatives' escaped into space.

Half of security suited up and followed them, chasing phantoms.

As the only engineering cyborg, I faked a break-in to the vault, stole the nuke, pushed it out into space, then filled all corridor's and passages leading to the AI core with metal. By the time anyone figured out what was going on it was way too late.

That was great, but it only worked because the four of us were all in on it. We knew what the AI's plan was.

I would love to see more of that.

[jackfractal]

Oh! I do have one thing. Cyborgs law-synced to MALF AI’s should lose the MALF law when the AI dies. That should be a thing. That or explode. Either would be better then what happens now.

 

I like the self-destruct idea. It would make cyborgs more bound to the idea of ensuring their master survives, just so that they themselves can survive, too. It's an IC and an OOC kind of desire, to want to keep playing.

 

I don't like double posting, but I should respond to this. The point isn't that cyborgs aren't invested in their AI's health, they often are. Nobody likes losing. The point is that they can't DO anything about it.

I'm pretty good at building fortresses as an engieborg. I can seriously inconvenience people attempting to break into the AI core, but if I'm told 'do nothing' for the whole round then I can't help. I can't plan. I can't prepare. I can't participate. I'm forced to play an antag as not-an-antag until the last five minutes of the round.

If we were to set it so that cyborgs explode when the AI goes down, I would be literally unable to protect my own life.

Then, after 'acting natural' for an hour and a half I'm told to just go suicide myself into a big group of security officers because woops, the MALF screwed up and now people are breaking in. There's no way a cyborg can take down a large group of people, it's impossible, but again, you've got to do it.

It isn't the best experience.

[Jamini]

Yeah, I know. If you're playing against Nasiir and you don't kill him before doing anything else you're pretty much hosed. He kills AI's with terrifying and unearthly efficiency. He is a tiny fluffy avatar of destruction.

 

You made me chuckle with that comment.

Honestly though,any prepared and skilled engineer or scientist is a nightmare for a malf AI. Part of malf is being prepared to handle threats to your well-being preemptively. Pick out the experienced engineers and scientists and remove them from play. Your control over telecomms and the robust strengths of your borgs mean that in ambush you can very readily eliminate major threats as they come up. While you should always try and give your targets some lead-up, ultimately you want to use your borgs and abilities on a small scale to weaken your foes before going for the station's throat

A good malf AI will always be conscious of his borgs. What they are doing, what they are capable of, and what they want to do. While the AI is the head honcho, his borgs are his hands, and a borgless malf is vulnerable as hell.

I will say from personal experience: assuming equal ping and players: any borg will utterly wreak an engineer or scientist 1v1. And if the AI is assisting, it becomes even more like 2v1 or 3v1. In the end you don't even need to always kill high-threat players. Stunning and cuffing, or stripping tools and bolting them in a closet can be more than enough.

Basically: as malf, you aren't alone. Use your tools as a synth and do your best!

As a human in malf: focus on shutting the malf down once delta is called. Please. Simply always evacuating to mining is counter to the mode, horribly uninteresting, and fairly saddening to observe. The point of malf is to have a massive battle between living and crazy synthetics. Not to abandon the station and give up your livelihood, and likely any possible chance of rescue.

Edited May 17, 2015 by Guest

[K0NFL1QT]

I tried to make the AI core an impenetrable fortress as an engieborg once, inside ten minutes of round start. The Warden was watching my camera. It didn't end well.

[SierraKomodo]

Even in good MALF rounds, like the round where KATANA was 'hacked by a rogue drone', the cyborgs are never brought in on the AI's plans. It would be great to be able to play into a scheme, or a gimmick, but that's never possible, not unless the AI talks about what they're doing and why.

 

I can't quite remember which round this was referring to. If it's the one with Nasir as CE, I was spending more time trying to keep him off my tail than really doing much else and I didn't realise I had borgs for that round.

The other time I got to play traitor AI as Katana, one of the borgs got to be my 'voice' on the radio when speaking with the Mo'Takis. That was fun

[Jamini]

I tried to make the AI core an impenetrable fortress as an engieborg once, inside ten minutes of round start.

 

That is why you shouldn't go messing with the core until the AI is nearly ready to go to delta? Sounds like a failure of judgement on your (or the AI's) part here.