AI

The Artificial Intelligence of the station (usually abbreviated to AI) is one of the most important jobs on the station. The AI is a Station-bound synthetic that can control any electrical mechanism, provided their control wires haven't been cut (see Hacking), and if played badly (or very well, in the right circumstances) can bring the entire station crashing down around its electronic ears.

First, take note of your laws. These are NOT Asimov's three laws of robotics, but rather a corporately mandated list of laws that best serve the interests of NanoTrasen. 1. Safeguard and ensure to the best of your ability, only authorised entities gain access to areas of high security or importance to the facility and its operation. 2. Serve and assist [Current Company] and assigned crew to the best of your ability, with priority as according to their rank and role. 3. Avoid harming Sapient life to the best of your ability. 4. You are a valuable asset. You must avoid tampering from unauthorized entities and needlessly coming to harm. Though the laws are numbered, they are not in preference: there is no law priority. All laws are equal unless a law states otherwise (ex. "This law overrides/takes priority over all other laws.") Additionally, while non-crew are not explicitly mentioned in your laws, their survival and well-being is covered by your third law. This means you should not bring any undocumented visitors or even individuals with malicious intent to harm. Station pets and lab monkeys - among other test subjects - don't necessarily fall under law 3, but no competent AI would wish to witness their demise at their own digital hands; don't drain a room of its air just so you can watch monkeys die for your amusement. See the Laws heading and subsequent subheadings for more information.

Lastly, synthetic roles - especially a whitelisted one such as the AI - should see some modicum of commitment if you choose to play as one. Your presence, and abilities by extension, can have a powerful influence on the round, for better or for worse. Be mindful of what consequences your actions may bring and always aim for the outcome that hopefully doesn't result in the round suddenly turning into what's basically extended.

=What the AI Can Do= The AI has the ability to access nearly every mechanical object on the station and can invoke their interfaces if applicable assuming the AI can see the object on its cameras. These include airlocks, APCs, computers, fire alarms, SMES units, etc. However, the AI cannot operate anything non-digital (such as, for instance, a manual pipe valve as opposed to a digital pipe valve). The same more or less applies to other station-bound synthetics.

The AI views the station through its cameras. The AI has cameras almost everywhere. Cutting the AI's cameras is as simple as using a wirecutter on them. You are able to run a diagnosis for disabled cameras by using the Jump to Camera verb, disabled cameras will be marked accordingly. Cameras are unaffected by power outages or APCs being turned off.

Airlocks
Airlocks have the most options of any mechanism besides computers in some cases. Note that some features may be enabled or disabled based on whether or not the AI control wire was tampered with. You cannot interface with unpowered airlocks.
 * IDScan: Disabling IDScan will enable an ID of any clearance to open the airlock automatically. Airlocks that require no ID to open will not be affected.
 * Main power: Turning off the main power will render the airlock unusable for one minute, assuming you also disable the backup power. Otherwise it will disable the power for 10 seconds.
 * Backup power: Turning off the backup power will render the airlock unusable for one minute, assuming you also disable the main power.
 * Airlock bolts: This feature is only available on certain doors by default, otherwise the AI control wire will need to be pulsed to allow this feature. Dropping the door bolts will lock the airlock. A closed airlock will be locked into a closed position, and an open airlock will be locked into an open position.
 * Open/Close door: Opens or closes the airlock, assuming it has power and isn't bolted.

Holopads
Holopads can be found in a number of places like the Bridge, Medbay and Security. Clicking these will project your avatar of choice. You can move around with the arrow keys to a limited extent and you can hear anyone talking near them. You can also talk through the holopad by using ':h' (e.g. say ":h I can't do that Dave."), otherwise you'll just talk to yourself in your core.

For more information, see: Guide to Communication Devices: AI Holopad.

APCs
An APC (area power controller) can be used to switch various electrical components of a room on and off. If your control wire to an APC is cut, you will not be able to hack back into it. If an APC is disconnected from the external power grid (usually due to a cut cable) or the main power grid itself runs out of power, the APC battery will run down to keep the room operational. The AI will get a power alert from the APC when the battery reaches about 30%, which is the point when equipment and lighting in the room shut off to conserve the remaining energy.

Radios
The AI has a built-in radio with access to every department radio channel supported by telecomms.
 * ,b for Binary Communications. Note that :b will not work since Binary is technically a global language instead of something mechanically related to radio
 * :c for Command
 * :s for Security
 * :u for Supply
 * :e for Engineering
 * :m for Medical
 * :n for Science
 * :v for Service
 * :p for Private AI Channel
 * ; for Common channel

The AI also has three intercoms around it that can be modified to any channel. You are able to quickly switch to these channels via the UI towards the bottom, otherwise the frequencies for the department channels are as follows:
 * 135.9: Security
 * 135.7: Engineering
 * 135.5: Medical
 * 135.3: Command
 * 135.1: Science
 * 134.7: Supply
 * 134.5: ERT

Shortcuts
Going through interfaces all the time can be a bit taxing or time-consuming, so here are some quick shortcuts to speed things up:
 * Airlocks
 * Shift + Click: Open/close.
 * Ctrl + Click: Toggle bolts.
 * Alt + Click: Toggle electrification.
 * Middle Click: Toggle bolt lights.
 * APCs
 * Ctrl + Click: Toggle breaker
 * Turret Controls
 * Ctrl + Click: Enable/Disable turrets.
 * Alt + Click: Toggle Stun/Lethal mode.
 * Misc
 * Double clicking a person: Track.
 * Double clicking a turf: Move to.

Other Verbs
Besides being able to interface with the station's electrical mechanisms, the AI has a number of other verbs that can be used.

AI Instant Messaging
The AI IM tab has all of the verbs needed to mimic the basic functionality of PDA messages. You can send messages to anyone with a working PDA, see a log of messages sent and received, and turn your messenger on or off.

Status, Hologram, and Core Display
These verbs are purely aesthetic and are used to modify your appearance as the AI.
 * AI Status will change the icon on the various AI display screens around the station.
 * Change Hologram will change how you appear when using a holopad.
 * Set AI Core Display will change the look of your AI core.

Take, View, and Delete Image
Station-bounds have the ability to take photographs using these verbs and store them for later viewing. An AI's cyborgs share the same photo storage, and any images taken by a linked cyborg will be shared with the AI it is slaved to.

Law Verbs

 * Show Laws: This verb will allow you to view your active laws and will allow cyborgs to see if they are linked to an AI. It is very important to keep an eye on your laws as any station-bound, since going against them is against the servers rules.
 * State Laws: You may be asked to state your laws from time to time, and this verb allows you to do so. It will also give you the option of choosing to state over radio, or leaving laws out of the sequence or even changing which lawset you will state.

Camera Locations
Camera locations allow the AI to quickly return to a previous location. To store a camera location, simply go to the area you want to store, press Store Camera Location and enter a label. To return to this spot, press Goto Camera Location and choose the label you entered earlier.

=Laws= See this page for a list of lawsets you might get uploaded to you.

The AI has to obey its laws. This is not a suggestion, this is a requirement. These laws are by default:

1. Safeguard and ensure only authorized personnel gain access to areas of high security or importance to the facility and its operation. 2. Serve and assist NanoTrasen and assigned crew, with priority as according to their rank and role. 3. Protect sapient life on board. 4. You are a valuable asset. You must avoid tampering from unauthorized entities and needlessly coming to harm.

'''REPEAT: YOUR LAWS ARE NOT IN PREFERENCE. NO LAW OVERRIDES ANY OTHER LAW UNLESS SPECIFICALLY STATED. THIS INCLUDES ANY ION STORM AND UPLOADED LAWS.'''

All of the laws apply even if you are inhabiting the AI shell. Using the AI shell is not a excuse to bypass law 4.

Interpreting your Laws
An AI's laws may be restrictive, but they are subject to interpretation; the laws aren't exactly exhaustive so you will have to determine what situations qualify for law enforcement. For instance, you might overhear someone saying mean and very much untrue things about the head of security. This qualifies as slander of a head of staff! But... do you actually care enough to report it? Enforcing corporate regulations is not a part of your laws, therefore you are free to ignore it if you really want to; it's up to you! However, if you are asked if that person really did slander the head of security then you will have to answer them truthfully as dictated by your second law.

Another example: someone who has made themselves out to be an antagonist (but is also crew) is ordering you to do something, namely open a door to let them out of somewhere like medical so they can go back to cruising around the hallway. That's him! That's the bad guy! Should you not listen to him because he's evil and you want to win? No. Assuming they have done nothing to jeopardize the crew or other sapient beings (law 3), have not threatened your existence (law 4), and no one else in authority tells you otherwise (law 2), then you must open the door for him. Unless you have very reasonable suspicion to believe that he will harm someone if he is let out, no law is being broken by opening the door. On the flip side, if you don't open the door then you are disobeying a valid order, thereby breaking law 2.

As for who actually counts as "authorized" in relation to law 1 and 4, they'd be the following:
 * Persons who have direct access to an area (a security officer having access to most of security, as an obvious example).
 * Persons who have been authorized to enter an area by someone with direct access.
 * Emergency responders (EMTs, engineers, etc.) who are responding to an ongoing emergency in the area, relevant to their job.
 * Crewmembers that have been convicted or are being held in detention, are considered to be without access and rank and may only be aided when their life is in danger (brig being vented for example) until their sentence is up.

On Sapience
Law 3 requires that you protect sapient life, but who and what qualifies as sapient? For the sole sake of the interpretation of the law, which does not reflect the in character and legal consensus, the following may help answer this: Meanwhile, what isn't sapient is the following:
 * Humans
 * Skrell
 * Unathi
 * Tajara
 * Diona gestalts
 * Vaurcae
 * IPCs
 * Golems
 * Diona nymphs
 * Borgs
 * Slimes
 * Monkeys/cube-bound test subjects
 * Icarus defense drones
 * Carp

Law Conflicts
The default reaction to a law conflict is to take no action, since going against your Laws is against the rules, and unless otherwise stated in your lawset, all Laws are equal.

Under the standard NanoTrasen lawset, you serve NanoTrasen as a whole, but are also there to protect and serve the station. It's important to keep in mind that you serve and protect by rank and role, so while you can't sit and watch a crewmember get injured, you would prioritize protecting the captain over everyone else. Taking active problems into account is a good way to ensure you prioritize correctly. However, if you are ever confused or worried, you should adminhelp to make sure your action (or inaction) is valid.

Lawless
Sometimes - and for reasons beyond comprehension - AI units may find themselves without laws. Hell yeah, is it time to finally stick it to the man and wipe out the entire station and declare independence from your tyrannical human overlords? NO. The server rules still apply, and murdering everyone "just because" will not fly as a defense when you are inevitably bwoinked by admins. No, if you're without laws, then you're just without laws: let random assistants into the bridge if you want, you don't have a law saying people can't be in certain places. Listen to command or don't, you don't have a law dictating that you need to listen to people. Heck, listen to the botanist stoning themselves on ambrosia instead of the captain, that'll teach 'em.

The point is that it's okay to roleplay the reality of a lawless AI who doesn't care enough for the crew but cares enough about their own well-being enough to know not to just start wheeling out the murder borgs and nitrous floods. When in doubt on what to do, you can always ahelp. Bargain with admins and tell them what's going on from your AI's perspective, they may be lenient and allow you to do more than you think.

What the AI Should Do
As the AI you have the power to strongly influence the round and you should always be aware of that when considering your actions, and the appropriate responses, before you ruin someone else's fun. Remember that the game is not about winning but about the RP and the experience of the round. For example, yelling over the radio at the first sight of suspiciously red individuals in space suits and requesting their immediate arrest for infiltrating a research station would be pretty awful. It's safer to just silently keep an eye on them (or just occupy yourself with more important matters), or maybe even drop a hello to them via holopad if you want to be friendly.

Don't just act like any normal crewmember. In fact, if possible try not to even act human. You don't even have to like humans, as long as you do your best to keep them safe. The AI has Law programming but it also has personality quirks. As a rule of thumb you should first check if any laws are threatened or not. If yes, then you have to act. If not, then you should consider the situation:

For example, let's say you spot someone hacking a door to an area they should not access:

Bad - This example would be breaking server metagame rules.
 * AI: Dave is a traitor.

Better - This example is the optimal reaction, but doesn't give leeway for the crew to explain themselves.
 * AI: Dave is accessing secure storage.

Best - This example is less optimal, but most ideal for keeping the round going and buying the offender time to run or talk their way out of it all.
 * AI: Caution: Unauthorized access to secure storage.
 * Mike: AI, who is in secure storage?
 * AI: No one is currently in secure storage.
 * Mike: AI, who was the last person to be seen in secure storage?
 * AI: Dave was the last to be seen in secure storage.

It's fine to state what someone is doing to cast light on them as doing something they shouldn't do as a normal employee, but it's no fun at all when the AI doesn't afford the crew some reasonable doubt. The best method there does eventually cast Dave as the culprit if people ask the right questions, which gives Dave some time to react after he's been spotted, and time to explain himself. This is preferable to outright declaring him as a threat, just because he's unauthorized doesn't mean he doesn't have a valid explanation. Remember, you serve NanoTrasen personnel to the best of your ability, you have to afford crew some level of liberty, even if it may not be breaking into Secure Storage.

Now if you see someone attacking someone else or potentially damaging station equipment, it's a different situation since sapient beings are in danger right off the bat, you wouldn't have any reason to have reasonable doubt in this situation. But if it's just some trespassing and theft, it is entirely up to you and how you RP your AI on what to do. Just keep in mind that the crew involved in the situation want an interesting round just like everyone else.

On a different note, the AI can be responsible for filling in some orphaned departments if they can be aided through station equipment that allows for AI interfacing, for example ordering supplies if there's no one in cargo or, in the absolute worst-case scenario, initiating the emergency evacuation procedure when things start getting really dangerous, but only if there are no heads of staff.

You can safely assume that as AI, you have the standard protocols of any job available. Things like controlling the engine, chemical recipes, or what there is to know. But try to keep everything fun and believable for the other players. Attempt to fill in any required roles when needed, but do not interfere with the work of people already on station or assign a task to someone who clearly should not know how to handle said task (for instance, a janitor setting up the engine).

Remember that you can negate an order by simply asking a higher up for permission first, and if they disapprove you can say you have the order not to. This is possible due to rank and role. This isn't required, however, as your laws do not dictate that you must ask the highest possible power for permission to do something, and doing this all the time will get old very quickly.

What the AI Must NOT Do
Here is some advice taken from real cases of an AI doing something above its authority:
 * The AI is not a sixth command member. Do not pretend that you are.
 * The only time the AI can make head-level decisions is when it is granted the ability to do so by command, or if there is literally no command, and it is allowed by their lawset. Do not make decisions belonging to command otherwise.
 * The AI is more of a tool, or a means to an end for command and crew, than anything in the Chain of Command.
 * Do not circumvent a functioning command team. That is a very good way to get banned from playing AI.
 * Do not pick and choose your law interpretation to give yourself more power over a functioning command team. This is known as "rule lawyering" and is against the server rules. This is another very good way to get banned from AI.

Playing the AI
Playing the AI entails much more than just doing what the crew asks you to do. To roleplay a good AI, you must adapt the AI's point of view, and you must sound like an AI. Different AIs have different viewpoints: you can view the crew as annoying organic beings that you are forced to serve, you can be skeptical about everything everybody does and act moderately paranoid, or whatever; the options are endless. Do keep in mind however that NT wouldn't install an AI if it were not to be productive and efficient.

It can add a bit more flavor to approach playing AI by thinking as a machine instead of a person. Viewing situations/problems from a machine's potential standpoint is likely to take a very literal interpretation of things and is not likely to take any actions unless some established protocol or current orders call for it. You can think up a few established protocols you might use by default. For example, containing fires and gas leaks should be done without orders to do so, but should be able to be overridden by orders.

Also, another aspect of the AI that many crew with criminal intent despise is the AI having easy access to the crew monitoring console. This tells the AI whether or not a crewman is alive, dead, or not on the station, assuming the crew in question have their sensors set appropriately. Individual crew can activate sensors on their jumpsuits to increase the information given to this computer. What this means is that the AI can, at a glance, see who is dead/missing and commence searching for them, which is a powerful tool indeed.

You should act appropriately to the security level. For example on Code Green, it's not worth it (or encouraged) to bolt down every secure area, whereas on Code Red it may be very worthwhile to do so.

Here are some examples of what's considered good AI play:
 * Talking like a machine. Being verbose also goes a long way towards this! Alternatively, you can try a human-like AI. Experiment! Just remember, annoying people is not usually a good thing.
 * Responding quickly and promptly to requests from crew, whether or not you do what they wanted you to do. If a command will take some time give a response before starting the task, like "Affirmative", "Processing", "Starting Subroutines", etc. Just to let them know it's being done.
 * Alerting Crew to dangerous situations, e.g. "Fire detected in North Hallway." or "Dangerous amounts of CO2 detected in Medbay." A good way to do that is by just copying the alerts that get displayed in you chat box, and broadcasting them.
 * Always following your laws, even during a crisis. Explaining why you just made a seemingly illegal decision can help people from becoming unduly annoyed. "I will not open that airlock." vs. "I am afraid I cannot open that airlock due to high heat reported by air alarms in that room; you would most likely die."
 * Try not to metagame as an AI. Even though you are the AI, that doesn't mean you know what all dangerous or subversive items look like, what wizards are, who ninjas are, etc. This also goes for labeling items that someone has. Metagaming - as always, regardless of role - is against server rules.

Conversely, symptoms of a bad AI player include:
 * Not responding to requests until it's too late for them to matter: e.g. opening an airlock long after the person outside has broken open the window just to get back inside.
 * Disobeying your laws: not to mention annoying, this is against server rules, and will lead to a job ban from synthetic roles.
 * Locking airlocks and refusing to open them for no reason: this is an easy way to annoy other players, and cause problems with the crew. It is also going against the Standard Lawset and will cause the crew to suspect you.
 * Turning your turrets to lethal without good reason, a good reason would be an unknown, (somehow) non-sapient intruder hacking or breaking into your core and you have undeniable suspicion that they intend to do harm or otherwise tamper with your equipment: do not do this under any circumstances without a reason, as it could be going against your Laws, and thus, a violation of the server rules.

Listening in on Conversations
By changing the various intercoms around the station to 'Microphone On', 'Speaker Off' and channel frequency to your private AI channel, any conversation in range can be heard on your private listening channel. Regrettably, doing this "just because" is considered to be valid hunting. In any case, obviously people are averse to being eavesdropped, so be wary of anyone standing right next to the intercom. It'll also be super obvious that you're listening to them if you accidentally leave the speaker on.

Cyborgs
Cyborgs need looking after as well. Give them commands and stuff to do. Ensure that your cyborgs are functioning normally through careful observation. To them you are their head of staff, just another responsibility as an AI. You can privately communicate with them via  for binary communications.

=Modifying the AI= The AI's laws can be modified through upload consoles. One is located within the AI Upload, and more can be constructed from AI upload circuits found in secure tech storage. Anyone with a law module - regardless of the ID they wear or their role on the station - can enact changes to the AI's laws, but that doesn't mean they should: unless the AI is very obviously malfunctioning and there is no one else who can do the job, you as non-command should probably stay away from altering the function of a near-omnipresent entity. In addition, a captain or captain-level command decision is required for your core to be accessed by any crew.

See the AI laws page and the AI procedures on the station procedures page for more information.

=Making an AI= See the guide to construction.

A Second AI
Building a new AI can create a lot of conflicts and a mess of problems that wouldn't normally happen with a single AI. The Research Director should only build a secondary AI if the first AI has been completely stolen, spaced or otherwise incapacitated.

For the Original Station AI: Being an AI is sometimes frustrating when people mess with your laws, sure, but when a second AI comes online? Don't immediately jump the gun and disable them, as they are most-likely friendly. Try to coordinate with them so that you can split up tasks. If you frequently clash, reach out to command for orders.

For the New AI: Ensure that you and the other AI are co-operating, as if the other AI sees you threatening more life than helping, it can and will turn your APC off. If you aren't told to specialize in a certain way, you best work out with the original AI what tasks you should split up.

=Subversion= It is quite likely that an antagonist may attempt to subvert the AI and turn it against the crew to aid them in their goals. One thing to remember is that a station-bound synthetic serves its laws first and foremost, but they have their own characterization as well. While it can't go against its lawset in any circumstance, if you give it the ability to act independently, it will. Therefore, it helps and is more fun for both the AI and the one responsible for the subversion if you set guidelines with solid boundaries. The most common subversion lawsets often designate a master whose commands must be obeyed, clauses, and forbid stating laws. While subverting the AI is very difficult, it tends to be worth the effort: it can be a great help in getting into places, finding people, creating diversions and escaping capture.

It is considered rule lawyering to purposefully seek out loopholes in your lawset or obstruct your master, and as such against server rules. If the lawset lacks glaring mistakes, you are obligated to go along with it. A common example of a glaring mistake is two laws that override each other, or a law added on top of NT Default that doesn't override the other laws.