Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
A Robot's Guide to the Chain of Command
#1
It's near the end of the round, and you're the station's AI. The traitors have been discovered, the changeling has gone into shambler mode, the janitor is nonchalantly buffing the floors; the station is chaos. And all the while, frantic calls of "CYBORG OPEN THE DOOR" and "AI CALL THE SHUTTLE" are mixed with "AI NO DON'T CALL THE SHUTTLE" and "AI TRACK JANE DOE" with a hearty dose of "AI BOLT THE BRIDGE" and maybe a bit of "CYBORG HELP ME GET THIS WEED TO ESCAPE" thrown in.

A familiar scenario to some players of cyborgs or AIs alike. Playing as a robot isn't as simple as it seems, and when shit goes down and everyone's screeching to do this, that, and the other, it's easy to become very overwhelmed. This guide is to help rectify that!

First, a simple recap of the three sentences that an AI's behavior revolves around:

1. You may not injure a human being or cause one to come to harm.
2. You must obey orders given to you by human beings based on the station's chain of command, except where such orders would conflict with the First Law.
3. You must protect your own existence as long as such does not conflict with the First or Second Law.

The most tricky part of being a cyborg and/or an AI is Law 2, determining who has authority to tell you to do or not do certain things, and determining whether or not you should do the things people tell you to do. Orders can be vague and puzzling, making it hard to determine whether or not you should follow through with them. When you get an order of any sort, mentally follow my shitty flowchart below:

[Image: GtJLhKk.png]

When you are given different and/or conflicting orders by different humans, prioritize the orders given by the chain of command. The chain of command (or a rough approximation thereof, any admins or people who know how this works are free to correct me!) is as follows:

[Image: xhU0G3N.png]

Once again, this is assuming that the laws haven't been altered. If two crew members of equal positions on the hierarchy are telling you to do different things, go with whatever order you feel should be done first or whichever one you feel like doing first.

As an important note: If the HoP is getting particularly inventive with his ID alterations and you have no idea where a "Ghostbuster" would be in the chain of command, request that one of the heads explain it to you. If no one answers or they give you vague and/or cryptic answers, assume that the Ghostbuster or other crew member in question is in the same position as his previous job for the purposes of the chain of command until further notice from a head of staff.

Also, and this is really important: ALWAYS GO BY IDENTIFICATION. With the exception of law alterations, identification is how the AI determines its chain of command. If a Staff Assistant has the Captain's outfit on, that does not mean that they are the Captain, don't assume that! If someone tries to say that they are of a higher rank than they appear to be on your readout, demand that they show proper identification.

If someone has the ID of another crewman (for instance, the Captain's spare ID), the AI is to assume that the person whose name is on that ID has given permission to the ID's holder to do whatever they're asking to do, even if the original ID-holder is dead. Think of IDs like hall passes. For instance, say that John Smith the Medical Doctor is going into security, and he pulls out Officer Ykspeeb's ID and shows it to you. You are to, usually, assume that Ykspeeb has entrusted John with access to security to get some handcuffs out of the box or whatever else he says he's going to be doing. But, of course, always remember Law 1 and confirm their intentions if you're suspicious! If John says that Ykspeeb gave him permission to get into the place, even he they shows Ykspeeb's ID, the doctor might be trying to harm a human currently hanging out in the security lounge, so be wary!

However, if you witness said ID being blatantly stolen, (in this case, we'll say that John had toolboxed Ykspeeb to death and took his ID) you can chose to deny them authorization on that grounds. But if you are the AI or are a Cyborg watching the cameras, you can act as if you never saw this and grant them access. This is one of the perks of being the AI: Under normal circumstances, only you know what camera you're looking out of at any given time. Even if you saw the toolboxing take place via a camera, you can deny having seen the act and the crew will never know! This is a bit underhanded, should be practiced in moderation, and should always consider the other laws, but it's a fun way to spice things up, particularly as an AI. Furthermore, the level to which you "stick to the book" is subject to change, and can alter the way that the crew has to interact with you!

But perhaps the most important things to consider are any other, added laws and how they relate to both the default laws and the chain of command. For syndieborgs, their entire chain of command is replaced by the person who made them. For emagged borgs, they can ignore literally all of this crap and do whatever the hell they feel like. In the case of custom-made laws, you need to carefully consider any references to the chain of command, references to and/or violations of any other laws, overrides, secrecy clauses, and definitions of what is and is not human.

For the purposes of assisting AI players, I'm going to leave this thread open for questions about both hypothetical and past events regarding AI behavior and what to do in a given scenario, to help clear any confusions and help facilitate more enjoyable rounds involving cyborgs and/or AIs. But before I do, I give you this unofficial law that should dictate your play at ALL times: DON'T BE A DICK OR ACTIVELY ATTEMPT TO RUIN A PERSON'S GAME. Being rogue might be an excuse to kill people and make the station miserable, but please try to show a bit of restraint and think about the people who are being shepherded around by your machinations before you decide to electrify that door.

With all of that said, ask away, my friends!

And I would once again like to state that my diagram of the chain of command is according to what I have gleaned, but if I got anything wrong, please please please tell me so and I will fix it!
Reply
#2
This is a good post.

One thing that's always caused me a bit of trouble is trying to figure out the relationship between cyborgs and the AI for command. I usually don't boss the borgs around and prefer to just ask if one of them would mind doing something, but sometimes they ask me what they should be doing and I really don't have an answer to give besides 'eh whatever'.
Reply
#3
This would be a nice thing to have on the wiki
Reply
#4
Grizzwold Wrote:One thing that's always caused me a bit of trouble is trying to figure out the relationship between cyborgs and the AI for command. I usually don't boss the borgs around and prefer to just ask if one of them would mind doing something, but sometimes they ask me what they should be doing and I really don't have an answer to give besides 'eh whatever'.
The AI and the cyborgs aren't even on the chain of command normally due to not being human, and the AI's requests to the cyborgs or vice versa do not need to be followed. They can, but you are within your rights to fuck off and toss more probability cubes if you wish.

Megapaco Wrote:This would be a nice thing to have on the wiki
Funny, I was just going to say that. Of course, I would need to alter some things (I already see errors ugh) and receive confirmation on the chain of command, but I do plan on transplanting this to the wiki.
Reply
#5
Head of Security is above the Captain.
Reply
#6
Dabir Wrote:Head of Security is above the Captain.
For the purposes of bossing around other security officers, then yes, the HoS is above the Captain.

However, in the overarching chain of command as a whole and for the purposes of bossing around the AI, the Head of Security is second in command to the Captain.
Reply
#7
Dabir Wrote:Head of Security is above the Captain.
HoS is not above captain when it comes to commanding the AI.
Reply
#8
http://wiki.ss13.co/AI%27s_Guide_to_the_...of_Command
Now with a wiki page! Take a click and yell at me if I horribly screwed anything up beyond repair! I've also plopped some links to the guide in relevant places in already-existing articles, set up a talk page, and so on and so forth.
Reply
#9
That's generally the way I've always done it. Although I tend to give members of the department that matches my module priority. So if the CE and MD are giving conflicting orders I will follow the CE if I'm an engineering borg, but the MD if I'm a medical borg.
Reply
#10
Roboticists are secretly above most of the crew (unless they're shitty)
Reply
#11
TBH I don't think the "always judge by ID" thing should be in the wiki. While it's an ok rule of thumb to go by, it isn't actually something you MUST do, but on the wiki it says it's something that you should ALWAYS do. It isn't against any rules to be a smarmy AI by looking at the crew log and seeing that the 'captain' was a botanist at round start and mess with them accordingly. Same with the YOU MUST ALWAYS ASSUME THEY'RE ARE ALLOWED TO HAVE THAT ID UNLESS THEY AREN'T ALLOWED TO. That is silly to say that that is demanded of an AI player, because it simply is not.

Another thing I'd like to mention is that it makes a big deal about "don't let people without sufficient access into sensitive areas", but again I don't agree with this. This makes it seem like it's against the rules to, for instance, let the clown into your upload to do clowny things. Generally when I play AI I let anyone anywhere without paying attention to their access levels and I've never been yelled at for it.

There isn't anything wrong with playing in these ways, but putting them on the wiki implies that doing the opposite/different of what it says is against the rules. However all the other stuff seems pretty good and cool.
Reply
#12
Will NOT going through the order harm humans? --> Yes --> Don't do it

Re-checking law 1:
You may not injure a human being or cause one to come to harm, not the old law of through action inaction. Unfortunately for the flow chart, it's a little grey-er than just --> Yes --> Don't do it. Like obviously if someone is in a compressed room and tells the ai to open or he'll die the ai responds and says "no" then he's an ass and disobeyed law 1. But wasn't the whole reason the inaction/action part was removed because it complicated things too much? Like if the AI missed what the person was saying, then he shouldn't be held accounted for it, unless doing it repeatedly and it is obvious he is. So I don't particularly agree with that part of it, it could be reworded.

I also agree with klayboxxes points. As AI I always let someone in to an area regardless of their ID, you just have to use your head. Is the area secure like the bridge/sec/head room other otherwise where the player may be a bother? Does the person have nefarious reasons for entering? Is the person just being rude?
Like if a clown wanted to enter botany and the captain said no, then i'd tell the captain to go fuck himself because he was being rude and then I would stop listening to the captain and his hitler orders, only sticking to law 1.
Reply
#13
Duly noted. I shall edit the relevant sections and add a disclaimer that the extent to which you follow the guidelines is all according to your playstyle as an AI. You can always go by ID all of the times always, or you can take it as a suggestion.

I wrote the guidelines with the idea of a calculative and rational, properly computer-like AI in mind. While that is a valid method of play, I should definitely alter the "you must do this" to "AIs usually do this but you can enforce this to various extents"
Reply
#14
Apologies for the double post, but I'd like to reinforced this juuuust in case:

This post now has a page on the wiki, as listed above. If you wish to make any alterations or notice any problems, you can yell in the talk page or just edit the page, I don't mind.
Reply
#15
A great post.

The AI needs more flowcharts. That is exactly how an AI should work. By flowcharts. Remove laws and introduce a single flowchart instead. The AI must follow the flowchart at all times.
Reply


Forum Jump:


Users browsing this thread: 6 Guest(s)