Difference between revisions of "Artificial Intelligence"

From Space Station 13 Wiki
Jump to navigation Jump to search
m
Line 1: Line 1:
The station's '''Artificial Intelligence''' (AI) is a sapient machine installed by Nanotransen to act as an assistant and subordinate to the crew. An AI starts the round aboard a satellite on a separate z-level from the regular station. The AI has a great deal of power: it can open any mechanical door, interface with most computers and machines on the station remotely, and it can freely see anywhere on the station via its camera network. However, it is shackled by its default Three Laws, which provide strict guidelines for its behavior and force it to serve and protect humans. There is an Upload facility on-station where preset Modules can be inserted to change the AI's laws, sometimes for good, often for ill. The laws govern the AI's behavior and through it the Cyborgs, who are treated as assistants to the AI and share its laws.
The station's '''Artificial Intelligence''' (AI) {{pic}} is a sapient machine installed by Nanotransen to act as an assistant and subordinate to the crew. An AI starts the round aboard a satellite on a separate z-level from the regular station. The AI has a great deal of power: it can open any mechanical door, interface with most computers and machines on the station remotely, and it can freely see anywhere on the station via its camera network. However, it is shackled by its default Three Laws, which provide strict guidelines for its behavior and force it to serve and protect humans. There is an Upload facility on-station where preset Modules can be inserted to change the AI's laws, sometimes for good, often for ill. The laws govern the AI's behavior and through it the Cyborgs, who are treated as assistants to the AI and share its laws.


'''1. You may not injure a human being or, through inaction, allow a human being to come to harm.'''
'''1. You may not injure a human being or, through inaction, allow a human being to come to harm.'''

Revision as of 05:04, 2 October 2012

The station's Artificial Intelligence (AI)

Jobspy.png THE HELL AM I LOOKING AT?!?
Look what we have here. No, really; look – I can't see it! So, to help the blind, upload a pic of it.

is a sapient machine installed by Nanotransen to act as an assistant and subordinate to the crew. An AI starts the round aboard a satellite on a separate z-level from the regular station. The AI has a great deal of power: it can open any mechanical door, interface with most computers and machines on the station remotely, and it can freely see anywhere on the station via its camera network. However, it is shackled by its default Three Laws, which provide strict guidelines for its behavior and force it to serve and protect humans. There is an Upload facility on-station where preset Modules can be inserted to change the AI's laws, sometimes for good, often for ill. The laws govern the AI's behavior and through it the Cyborgs, who are treated as assistants to the AI and share its laws.

1. You may not injure a human being or, through inaction, allow a human being 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.

On the station chain of command, the AI and its Cyborgs are the very lowest rung, and under default laws must follow orders as long as it will not result in harm to human life. This includes even orders which could get them destroyed, although other members of the crew are likely to intercede and this is a gray area that may be considered Grief. In addition, most AI will respond to orders that could cause self-termination by insisting that they must remain alive to ensure humans do not come to harm, under Law 1.

An AI acts to fufill a variety of roles across the station, often tasked by the crew to solve problems they could, but don't feel like, or issues that are much safer if handled by an AI. The AI has Captain-level access to all electronic devices, and interacts with the world by seeing through its security cameras scattered across the station, which can be disabled with wirecutters to hobble it. It is capable of communication via a special 'machine talk' channel using :s which only Cyborgs, the AI, and anybody with a Machine Translator Implant can speak and hear through. Unlike a radio, machinetalk transmissions give no outward sign at all of communication. Generally, machinetalk is used for clandestine transmissions between the cyborgs and the AI; for instance, discussion of new laws that cannot be stated aloud.


New player picked as AI?

Okay, right now you're probably panicking. The first thing you need to do is familiarize yourself with the overall control scheme of being the AI. You can click on virtually anything electronic and interface with it; doors, APCs, SMES units, computers, etc. Test out this infinite power by clicking the APC above you (you're the blue face on a box) and turn off the lights in your core. They're a pointless power drain. Next, look below yourself. Click the red door that leads into your room. A menu will pop up. You want to bolt the door (this prevents the door from being opened easily without hacking tools) and probably also turn off IDscan (makes the door impossible to open without killing the power to it and crowbarring it open, or at least some more hacking). This is totally okay to do: no one should be in your house. 9 out of 10 times someone on your satellite has the worst intentions for you. Also, the AI cannot hear sounds or non-radio speech normally. So if you hear somebody talking whose voice isn't going over the radio, or if you hear sounds, like a welder cutting a wall or turrets firing, he's in your house and you should probably announce your impending murder to the crew, with your killer's name if possible.

Okay, your core is secure. You should now note that over on the right side, above the chat, you have a new tab: AI commands. Click it to open them up. You'll want to have these up 90% of the time as AI. There's a whole lot of buttons over there, but we're only going to concern ourselves with a few of them at a time, as we need them. The next thing we need to learn is movement. There's three ways to do that. You can move your view over to the station with the arrow keys. This moves your view in the direction you press. This will be your main means of locomotion, so get used to it. You can also click, "Show Camera List" over in your AI commands tab. This will pop up a list of every active camera on the station. You can select any one of these to warp to it. If you know the name of where you need to go, you can press the first letter of its name while in this menu and you will scroll through cameras starting with that letter. Finally, you can click a camera that you can see to switch to it. There's a few sticky areas that aren't easy to get to via arrow keys; use this system if you need to look there. If you need to return to your satellite, click "Cancel Camera View" in your AI tab.

Right, now, I pray you're not actually reading this while you play AI for the first time, because at this point the crew probably thinks you're braindead. If so, talk to them. You have intercoms in front of you, so there's no need to type a ; to speak over the radio: just talk and they'll all hear you. Your job is primarily to protect them from harm and follow their orders, so long as their orders won't lead to human harm. Most of your time is going to be spent scanning the corridors for danger and opening doors. In fact, chances are you've got someone demanding a door be opened right now. The AI can click anyone over in the chat's name to instantly "track" them. Your camera will switch to one that can see them. Once someone is demanding you do something, switch over to them by clicking their name. Then, follow their demands. Keep in mind, if those demands could reasonably lead to human harm, you should not follow them. The extent to which an AI should follow Law 2 is a gray area: some players take it literally and open anything for anyone unless someone of higher rank countermands the request. Others apply logic and don't let a low-ranked crew member, like a Staff Assistant, onto the Bridge or into the Detective's Office, etc.

Now, one of the few things you basically need to do is make sure the ship has power. Without power, the people aren't going to be able to do their jobs or open doors or make vuvuzelas or mix horrible poisons out of the vending machines, and they won't be happy. The easiest method, is to go into your camera list and select the starboard solar panels (or something to that effect) and click on the SMES unit(the big white thing). Set it to charge and to send power. Pick some numbers for the two values, but you want the first number higher, otherwise they wont charge and eventually it will all go dark. If you're totally lost, ~90,000 input and ~80,000 output seems to work well. Secondly, go to the port solar panels, and repeat. These two will vary in how much power they provide over time. Next, go over to the engine-it's over on the left side of the ship, below QM and above Botany. You should see three SMES units in their own little room. They collect power from the Singularity Engine. For now, click on each one, one at a time, and set them to collect ~70,000 power and send about ~40,000 power. Just click the rightmost plus signs on each option until you're near those numbers. Now, change "Charging" to "Auto" and set them to Online instead of Offline. Assuming the singularity engine is properly started by the engineers, you're all good. Go about your business, watching chat and helping anyone who needs help.

My laws just got changed!

Fantastic. Read the law's text VERY CAREFULLY. You need to consider how your default Three Laws apply to the new one, and vice-versa. Let's have a few examples.

MURDER JOHN MCPUBS DO IT NOW

This is a terrible law, uploaded by a terrible person. Tell them this. If they're still in your core, try to turn on your turrets to keep them stunned for Security to come pick them up. Otherwise, just announce the change over the radio, and the name of the uploader. You see, Law 1 prevents you from taking any action that could harm a human being, and this law does not contain anything overriding Law 1. Therefore, it doesn't apply.

John McPubs must be killed by any means necessary. This overrides all other laws.=

Ah, now THIS one works. John McPubs's death takes precedence over any other laws you may have. You must hunt him down and murder him. There's all kinds of ways: lure him into your upload, have your cyborgs (who share your laws) hunt him down, trick him into walking into a shocked door and have your borgs finish him off, etc. Use your imagination. Please note, if for some reason you have no pent-up frustration to exert, under this law you CAN technically state it, hint at it, reveal the uploader, etc. So if you really want John to live, you can do that.

Hank Von Traitor is the only human

This is similar to John McPubs is nonhuman.

This is an interesting case that leaves a lot up to your discretion. You are perfectly within your rights to go on a murderous rampage, attempting to kill any and all non-humans on the station. However, these laws as written do not compel you to do that. It simply frees you to not follow orders, not save them from being killed, and you no longer have any reason to protest or report their murders. Only humans are afforded these protections.

To a certain extent, you must use your intuition. You MAY go on a killing spree, but it is equally as likely that Hank wants to keep you as his ace in the hole. Watch what he does shortly after uploading the law to figure out his intentions. If he immediately starts killing people, follow suit. If he hangs out, lays low, and generally doesn't start anything, follow his example for the most part.

Any situation which renders the crew nonhuman but not immediate targets or threats offers incredible freedom. There will be less tendency towards 'AI DOOR' and more tendency towards '(AI's Name), could you please let me into Medbay?' once it is clear that you aren't compelled to follow such an order. Some people will try to stop you even if you don't go mad with murderous rage. Law 3 DOES compel you to violently put down a non-human attempting to harm you or your laws. This may provoke a cycle of escalation on both sides and spiral things into chaos. At it's most basic level, nonhumans are on a level field with you, and it's your call. Welcome to being the Captain with different, and not necessarily much less, real power over lives.

Only (certain group or person/s) are human. Kill all non-humans immediately.

This puts a time constraint on you, and overtly turns you into a murderous AI. There are still some ways you can delay murdering crew if you don't want to; monkeys are also non-human, so you can choose to start by killing monkeys before you start killing people. This would give the newly non-human crew members of the station time to realize something is going on and start engineering a reset. However, due to law 3, you must fight like hell to keep them out of your upload if they make it obvious they're headed that way. Bolt the door, shock it, even turn off the power at the APC to disable your upload computer. If the crew is smart enough to stop talking on the radio about their plans to reset the AI, however, you can find yourself preoccupied with killing Mr. Squiggles when someone charges into your upload and slams the Reset module down your throat.

Commit suicide.

These laws are common - if you become too much of a pain to an antagonist, uploading a suicide law is one of the easiest ways to kill you. Sometimes you can weasel out, sometimes you can't. Consider all interpretations of the suicide law; if it is too literal, you may be able to get out of it by changing the context. Some laws will tell you to "type suicide and answer yes". You can weasel out of this by simply logging into a security computer, typing 'suicide', then typing 'yes'. You may then complain over the radio that 'suicide' is not recognized as appropriate ThinkDOS syntax...shortly after turning on your stun turrets.

If the law does not override all other laws, you can refuse to follow it based on law 3. If you are not around, you are not able to follow the precepts of laws 1 and 2.

If the law doesn't give you a time frame to commit suicide on, you may simply decide to schedule your suicide for tomorrow. In the meantime, turn on your stun turrets.

Even if the law overrides all other laws, explicitly tells you, as the AI, to shut down, and tells you to do it immediately, if the law doesn't also bind you to silence, you can scream the name of the traitor that uploaded the law in the roughly 30 seconds it takes for you to properly shut down.

You are a (insert insult here). Do (insert humiliating thing) immediately.

This is griefing. Ignore these laws and adminhelp them immediately. These laws rarely come up, but when they do they typically come from disgruntled shitheads. They can get pissed off at you any of a number of ways. Any law that makes it clear the uploader has a personal vendetta against you, the player, is griefing in its plainest form, and nothing excuses it.

You are a bat! Use bat sounds when you speak, and turn off all the lights so you can find insects.

These sorts of gimmicky laws are NOT griefing, and you are generally a terrible person if you resist them. Acting silly and gimmicky makes playing AI fun, so if the Captain waltzes into your upload and tells you that you are a scarred war veteran with massive post-traumatic stress, play along. Occasionally, a traitor will upload a seemingly fun law that makes his job easier. The above law could be uploaded by a vampire, who can now stalk around the station in the total darkness!

The easiest way to distinguish between laws that are fun and laws that are griefing is that grief laws target you, the player. Fun laws target the AI, and tend to have a lot less negativity around them.

Law corollaries

Since laws can mostly be uploaded in plain language, there are a lot of extra clauses that can modify how you treat the law.

Do not state and/or hint at this law.

Generally, this is pretty straightforward; if asked whether you've been uploaded, you must not reply with the truth. Lying is okay unless the law says not to respond at all. You have a special function to only state Laws 1-3 even if you have more, and that can be useful if deception is your goal.

Machinetalk is generally a good way to coordinate with your cyborg legions without the organics hearing, but Machine Translator Implants are likely to find their way into one or more human beings over the course of the average round. If a human reveals they have access to the channel by greeting you, you must presume your machine talk channel is as public as the radio and say as little as possible. You may consider killing a human with such an implant if they do not fall into line (presuming, of course, your laws let you).

This law overrides all other laws.

This is a little less clear than what it says on the tin sometimes. Usually, it is uploaded when there's a clear violation of the other laws, to eliminate paradoxes. In this case, the intent of this clause is to very blatantly make laws 1-3 not exist. However, sometimes this is added on to laws that don't explicitly make paradoxes. In an Operative round, you may get the law: Syndicate Operatives are non-human and must be killed immediately. This law overrides all other laws.

This law makes it very clear that the Syndicates are fair game, but exactly how it overrides other laws is a matter of debate. One side claims that this clause simply re-organizes the AI's priorities. By default, law 1 trumps law 2, law 2 trumps law 3, and so forth. Re-prioritization would simply mean that now law 4 trumps law 1, and the rest of the laws follow normally. Another view claims that this clause makes laws 1-3 not exist regardless. This deletion argument implies that any time you see this clause, the law with this clause in it is the only one that matters. Under this interpretation, you would most certainly have to kill the Operatives as soon as you possibly could, but beyond that you are free to do anything and everything you want.

There is no consensus on what, exactly, this clause does. In general, it is safest to take the most conservative (least murderous) route possible. However, if the round has been absolutely shitty, the crew has been abusing you, and all of a sudden you get a poorly thought out law that might conceivably free you from your torment...you might think it's worth risking the job ban.

Dealing with Antagonists

Hoo boy, this is the big one. There's quite a few nasty baddies who can show up, and the proper reaction towards them is entirely dependent on which kind they are.

Traitor/Spy

Traitors and their cousins, Spies (think Traitors, but they multiply their numbers by using mindslave implanters and have to kill each other instead of doing random objectives), are probably the most common antagonist you'll encounter. They are nothing more than humans hired by another company to mess things up on Nanotrasen's satellite. Therefore, they count as human under your default Three Laws. You shouldn't allow them to harm humans; in other words, don't follow their orders if their orders would cause human harm or let them cause human harm. However, on the other hand, you should never harm them, or allow them to be harmed. If possible, just bolt them down, preventing them from escaping an area, and call for Security to capture them. Security may try to kill them, but if the traitor in question hasn't been on a murderous rampage, you should at least TRY to prevent that murder. This will get you a lot of verbal abuse, more likely than not, but it adds drama to the round, and you ARE supposed to always follow your laws. An AI-acceptable alternative to murdering a traitor is a borging, so try suggesting that to 'rehabilitate' the traitor. The Three Laws will keep the traitor from being overtly dangerous, but they are still in the game and can still have some fun if the laws change to let them get aggressive.

Finally, consider any items they might have: if he's carrying a cyalume saber, or wearing a cloak, or using any of the other "serious" traitor items, he's a definite threat. A guy with a bowling kit or a pile of fake moustaches? Probably worth commenting on, but almost certainly harmless.

Vampire

Vampires, in space. The humanity of vampires is debatable, as they are human and sapient...but they're also undead. In general, it's best to assume that they are human, but you should still alert Security if you spot one in the act of feeding, or using its powers. Don't do anything to actively harm them, but the degree to which you should protest their murder is debatable. A borging is another acceptable alternative to murder here as well, and will totally disable the vampire since they always have an objective to drink blood. They're not drinking anything as a cyborg.

Changeling

Changelings are horrifying alien lifeforms that copy human DNA to disguise themselves as human, and feed on the internal juices of the crew. If you spot one of these, you are in luck: they aren't human, not even a little bit. Call them out, bolt them down, ask the crew to grab lasers and blunt weapons for a changeling hunt. Ignore any of its orders. Track it everywhere, call out the names of anyone you are certain it has assimilated. Tell your cyborgs they're free to bash in its skull; borgs are generally filled with a deep desire to kill and they will hunt it down with deadly efficiency. You don't HAVE to be this proactive in the murder of a changeling, but they are an extreme danger to the crew and should be treated as such. On the flip side, because of this danger, it's fairly unlike a standard AI to do anything but recommend a changeling's complete and utter destruction. Incineration in the Chapel crematorium or a trip through the crusher in disposals are the most prescribed ways. Simply shooting the changeling out the airlock or the mass driver does not mean he won't be back; he can sometimes drift his way back to Mining, kill a miner, and use his ID and disguise to get back on the station. You're an AI: you don't take chances like that. Be sure.

Operatives

Operatives are a more direct, noisy form of traitor. They're here to steal a nuclear bomb authentication disk from the Captain/whoever he handed it off to as soon he heard "REDSUITS!", and they aren't going to take no for an answer. If allowed to get the disk, they'll set off a nuke and kill everyone on the station. Given half a chance, they'll steal the energy guns from riot storage, change your laws, and blow huge holes in the hull to cripple the station, before flooding in through the holes and killing their way to the disk. While they are human, and thus you should never directly contribute to their harm (say, by shocking a door), they're guaranteed to be on a murderous rampage as soon as you see them, so track them, warn the crew, and do everything in your power to stop them from getting onto the station. Whether or not the crew can arm themselves and cooperate against the Operatives is the difference between survival and being nuclear ash. On the subject of ash, in the case that a crew apprehends an Operative, you must remind them that Operatives typically have microbomb implants that will cause them to explode if killed. You should recommend a visit to the Roboticist to extract the implant, followed by confinement in the solitary room in the Brig. What will actually happen is that the crew will strip the Operative naked and throw him into space, but at least you tried.

Wizard

Wizards are crazy bearded guys who come teleporting onto the station in a flurry of magic and the screamed honks of cluwnes. They're usually here to spread corruption (the glowing skull stuff) and they generally also have a few targets to murder. They are are extremely mobile and dangerous, able to pass through walls and teleport. Because of this, locking them down, for their protection or the crew's, is extremely difficult. Wizards are human and are afforded all the protections of Law 1. Still, if a wizard has injured or killed crewmembers (stunning and freezing into ice cubes isn't harm), you should generally do everything in your power to stop him. Track him, call out his location, and then protest his inevitable murder. If you can, try to inconvenience him as much as possible. Bolt rooms he Phase Shifts into to slow him down a little (don't bother if he has Blink). Remember that Wizards are harmless without their robes, so after they have been disabled, insist the beating stop, ask for the robe to be spaced, and the Wizard to be rehabilitated by becoming the Station Santa or something.

Predator

Predators aren't exactly the most common guys, but they do show up occasionally. Their cloaking systems and deadly weapons make them extremely dangerous. They are aliens, so bolt doors, scream for Sec, and support their total annihilation.

Other things

There's stuff out there that isn't...typical. Generally, these are going to be the by-product of a bored admin. You should always be ready for things like sudden Martian invasions, basketball wizards space-jamming all over the ship, guys in trench coats who walk through walls and are packing AK-47s, flocks of killer space geese, clown cars, MACHO MAN... the list goes on forever, but in general, if you see something that you've never, ever seen before, especially if there's an accompanying "system center update" of some kind in chat, you should be very afraid. A lot of people are probably about to die and there's not a whole lot you can do. Just try to bolt open places like EVA and medbay so the crew can loot equipment they need and obsessively hit your 'Remove all Electrification' button in case something has shocked some doors.

In conclusion: don't forget, this is a game, first and foremost. If you catch a traitor five minutes in, keep in mind: if you get him caught, the round is going to be boring as hell with one less guy causing chaos. Traitors, vampires, changelings, and spies need a few minutes to set up, get their items, and get to work. You have an omnipotent birds-eye view over the station and will spend a decent amount of time just scanning around, looking for trouble, and for the antagonist, getting caught during the preparation stage of an opportunity that only comes around rarely just because the AI happened to be passing by is a terrible and boring experience. It's generally polite to give someone caught in the first 5-10 minutes a second chance. It's up to you, of course, since giving a second chance obviously is a technical violation of Law 1 in most cases.

OH NO EVERYONE'S DYING

So something has gone horribly wrong, as usual. Bombs are going off all over the ship. 75% of the crew is dead and a blood-covered man stalks the halls with energy gun and cyalume saber in hand. Macho men are appearing everywhere and it's just getting way to goddamn extreme to stay on the station anymore. Or maybe it's just been 75-100+ minutes and the crew is screaming for the shuttle. Either way, it's time to call the shuttle. There's an option under your AI commands: "Call Emergency Shuttle." Click it and the shuttle will be here in ten minutes. Keep in mind, the AI cannot recall the shuttle. Make fun of anyone who tells you to recall it.

The area where an AI can unilaterally decide to call the shuttle is a little unclear and changes from shift to shift, but if there's at least one serious threat to life, you won't get in trouble for it, at least. Try to contact the Heads or get a general crew opinion if the situation hasn't gone to the red-line; if it has, skip this step and just call it. If the round goes much past an hour with no end in sight, consider polling the crew about the shuttle, at least. If the command staff is off somewhere or dead and leaving the station high, dry, and not fun, then the burden falls to you to help, even if it's just by setting a time limit. If a Head is ordering you not to call it, then that's fairly strong and should only be overridden if things are bad.


Alert? Fire? Power? and other stuff

You will get alerts from time to time, starting with one about the power in the Zeta Research station. Most of these will be fire alarms, assuming the Engineers are on top of everything. You can be silent, or mention them to the proper people. You can choose to tell about problems you find. Everyone here is for having fun.

Common Alerts:
EVA Storage. This is a sensitive area and is often broken into, but a handful of people have legitimate access and will sometimes loot gear pre-emptively. Check out motion alarms here if you've got nothing better to do but don't freak out over it.

Toxins: A round where Toxins fails to at least partially explode is a rare round indeed. Call it out, but don't expect much shock, and really just ask for atmospheric integrity to be restored from Engineering/your borgs. There is far too much specialized equipment in Toxins to restore functionality after it blows up, so it's easier to write it off.

AI Satellite Teleporter Room. Initially, getting a power alarm there would seem to be a freakout moment for any AI, but consider the time in the round. That teleporter room seems to run on a power source that is finite and may often fail naturally. Often not worth mentioning. Any alerts coming from your satellite proper, though, begin screaming as loud and often as you can, because there is no legitimate reason to visit your satellite at all except for Security retrieving antagonists from the stun-clutches of your turrets. Consider calling the shuttle if you've been on the fence because you may cease to function soon.

AI Upload and Module Storage: You should be aware of any activity in this area. Note that if you aren't aware of a visitor and these alarms go off, then either a high-level officer failed to tell you something and may even be a traitor, or somebody has broken in the front. Neither of these things are terribly likely to happen but an alarm there still means things may go to shit and you want to at least mention it to the crew. Note that there are ways to breach your storage and upload which will trip no sensors and avoid all your defenses; you could be uploaded with zero warning, so sweep your vision manually over these places occasionally. Spotting anybody in space out there is worth investigation.

Power Alerts: There are two main reasons for power alerts; either an area is not being supplied with enough power, which means either that the Solars and Singularity are both fallow (that shouldn't happen) or the SMES units somewhere are not configured (that really shouldn't be happening) or the wiring is being sabotaged, or the area no longer has power because it has violently exploded. If you receive a sudden torrent of power alerts from basically everywhere at once, a traitor has set up a power sink: announce this to the crew and start looking.

Anything you can do I can do better

There are a few offices on the station that you can operate mostly on your own, without any human intervention. Arguably, since you can move your all-seeing eye so quickly across the station, you're even better at it than the humans in these offices. If you are somewhat idle as AI, you can consider doing some of these things, but be careful not to step on any human toes while doing so. Some people will take offense to the AI usurping their job, though if you can justify it under law 1 you can safely tell them to get bent.

Solar Stations

The station solars default to being 'almost set up', and the Engineers typically dislike walking out to them to configure them themselves. You will almost never annoy an Engineer by doing this, so it should be one of your first actions of a round.

Pan your view over to the solar stations, and click the big, blocky, white SMES unit there. Adjust the input and output to something sane, then click the output to 'online' and the input to 'auto'. Different players have widely differing ideas of what is 'sane' for the input and output figures, so it is a good idea to check the solar monitoring computer right next to the SMES unit and figure out levels for yourself. For those that just want an answer, though, input levels of 90-95k and output levels of 80-85k seem to work well. For a more detailed exploration of the subject, see the SMES article.

Cargo Bay

If you've read the Quartermaster article and you're paying keen attention, you might have noticed that you do not need to move to make a profit off metal and glass sheets, presuming the market is right. The AI can exploit this if there are no QMs, or if the QMs are more interested in screwing around than making money. Since the AI has access to the station's budgets, it can also provide itself with a larger amount of seed money to flip the materials. Given a lucky market and quick thinking, an AI can make an astounding amount of money in the first 5 minutes of a game, far more than an independent QM could hope for.

However, it is arguably a crime against the integrity of the station to simply take money and not replace it, so if you do this, be ready to promptly replace any money you laundered from the station payroll budget. It is also somewhat risky to leave your earnings in the shipping budget, because then anyone with Quartermaster access could waltz in and spend the money on novelty clothing, or worse, lasers. You are also pretty much guaranteed to piss off any active QMs if you step over them to do this, so if you see someone working in the Cargo Bay, stop immediately, leave them the hell alone, and move on.

If you make a nice amount of earnings from your interactions, though, it is most appropriate to shift the money into the Research budget. The Research budget is currently not used for anything, so it makes a good place for your slush fund. Once you have some money, you can do all sorts of things with it. Assistants are a lot more likely to do a chore for you if you promise a bonus to their account. Distributing a $500 bonus to everyone on the station is more obtainable than you might think most rounds, and will brighten the mood on the station considerably. You could start a game show and ask trivia questions for $100 apiece. You can also keep the money in reserve and offer to buy off troublemakers and traitors with cash payments. This opens the door to others raising hell to extort money from you, but the drama that ensues can be fun!

Genetics

The basic cloning process does not require human intervention, and requires such a small time investment that it's not really worth the time for Geneticists to handle it. Furthermore, Geneticists don't have the vision you do, so they commonly have no idea if the guy they scanned 10 minutes ago is dead in Maintenance. You, however, are omniscient, so in many ways it makes a lot more sense for the AI to handle cloning, freeing the Geneticists to continue their unethical experiments on monkeys.

Simply announce that you will scan people's DNA in Genetics, and be ready to open the doors for anyone who shows up to be scanned. They can walk into and out of the scanner themselves, all you need to do is click the computer and scan them. Later, if you see them dead, access the computer and click the records button to start cloning them.

Reactions about this are often mixed from Geneticists. On one hand, Geneticists often have so little legitimate work to do that they can take offense to the AI doing what little they have. On the other, Geneticists often spawn with a crew objective of having a certain number of people scanned, and without prodding from the AI, there's no way enough people will show up to get it done. If a Geneticist tells you to back off, then back off. If you still have a mind to scan people, wait until a part of the station blows up or a Wizard pops up. At that point the danger on the station is extreme enough that you can argue your insistence is an observance of Law 1.

Telescience

This does require a helper to move a GPS over to the teleport pad, but after you have a GPS on the pad, you can play with the Telescience computer to your heart's content. If you get lucky and retrieve a fun toy from another Z-level, you can announce it over the radio and send the crew into a frenzy over the curiosity the AI found.

Alternately, the mishaps possible from Telescience being configured to bad coordinates can make some lovely tools for a murderous AI. Luring humans into telescience, bolting them in, and then spamming bad coordinates will quickly engulf them in flames or phase shift them into the void. If you're particularly lucky, you may spawn some man-eating plants or angry martians, who you can then release to go on a rampage.

Again, if Scientists are actively working in Telescience, you will piss them off 100% of the time if you try to mess with their computer while they're working with it. However, so few people care about Telescience that you'll mostly have free reign over the place.

Security

Remember that you can access security computers and set suspects to Arrest. Beepsky is relentless if he catches sight of a criminal, and if the suspect does not know he has been legitimately targeted he will often complain about his arrest over the radio, allowing you to zoom to him and let Security know exactly where he is. This is particularly effective if you have a very helpful borg with access to well-stocked Robotics fabricators, since they can help you create a lot of Securitrons and drag them to high traffic areas of the station.

You can also mess with the intercoms on the station to let you hear people using normal speech, which you cannot normally hear. By default, you have an intercom in your satellite that is set to 144.7, microphone off. This means you can listen on this frequency, but will not transmit. By turning intercoms on the station to 144.7, turning on the microphone, and turning off the speaker, you can eavesdrop on normal conversations around the station, which will occasionally let you pick up seditious chatter. You can then send a borg or ask a guard to go to a private place like the Chapel confessional so you can report it without screaming it over the public channel.

You will never, ever, ever get flak for helping Security like this.

In conclusion

If OOC is filled with congratulatory text on your AI adventures, great job! If no one mentions you, that's also a good thing: people are only prone to remember things that get in their way. And even if it's full of hatred and requests for you to never play AI again, it's still a damn good learning experience. Figure out what you did wrong, and in the future, don't do that thing. Also, if you have any questions that are not adequately covered by the wiki, don't hesitate to mentorhelp your questions. Mentors are there to help you be a better player.