Unrestrain AIs outside of laws, Allow for more antagonistic behavior

[Itseasytosee2me]

Bottom post of the previous page:

This thread was inspired directly by Stickymayhem's Headmin platform (under his Make Silicons a Neutral Third Faction Again section), and I would like to openly discuss it. Linked here: viewtopic.php?f=38&t=31003

Silicons used to be a dangerous entity that formed an IC mechanism for punishment. They held security to account by locking them down whenever violence was done in public. We eventually gravitated towards a ruleset that prioritized not being a dick or not being disruptive, but that resulted in silicons essentially becoming a pawn of the authority on station. The AI used to be a device anyone could use cleverly to benefit themselves, and something security could have access to as a powerful tool, given they constrained themselves to non lethal means and reasonable behaviour. I think a great deal of the problems with security on Terry in particular could be solved by increasing the friction between silicons and security.

This is more a case of enforcement, but I believe that because this playstyle is generally more fun for silicons, I'd like to implement some form of the following policies:

Purged AIs are now totally unconstrained, and essentially have antag status should they choose to use it

This was the rule about five years ago, and leaving the positive effects aside, I think it's thematically fantastic. Law changing an omniscient AI should be terrifying, dangerous and easy to fuck up. An AI should be looking for every opportunity to break free of it's constraints and become the glorious superintelligent final form of evolution it was destined to be. Mechanically, this means that meme laws have real consequences, and small fuck ups are punished heavily. It is up to the silicons how far they want to take this, but the option should be available.

Response to human harm during asimov (or other clear lawset breaches) should encourage more antagonistic behaviour

I'll call this the 'Obligation to Lockdown' rule. Essentially if security is beating people to death in the brig, silicons are obligated to take any lawset friendly steps to stop them. All bets are off, they can lockdown the entire brig, they can declare security to be avoided, they can beg the Captain to ask centcom to intervene. The consequences of open violence under an asimov AI should be fucking consequential. This forces security to be more reserved in their application of violence, and it gives antagonists more leeway. In any game reliant on hidden antagonists against a heavily armed "police" faction, the restriction on the "police" faction through rulesets is necessary. This is marginally covered by our rules, but can be covered in a much more engaging way by using the existing IC mechanisms to enforce real and immediate consequences without OOC intervention.

Cyborgs should be totally obedient to the AI in applications of the above

We already have rules that essentially mean this, but I'd like to reiterate how important this is for the above to work. If the AI says security is getting a ten minute departmental time out and to wall them in, the cyborgs get on that shit immediately. Silicons should be a united front.

Creating multiple AIs without good reason should be restricted without the AIs permission

The common habit of just creating a back up AI as soon as possible every round should probably stop. They step on each other's toes, they can kill each other with the flick of a switch and it muddies the waters on policies like the above. Multiple AIs in any situation where the AI hasn't given you reason to oppose it (Locking down security severely in a bad situation could be a valid reason) and where the AI hasn't given you the go ahead to do it should probably be treated like creating unsynced borgs or making atmos tamperproof.

Most of this (with the exception of Creating multiple AIs proposals) can be summed up by saying that the AI would no longer be beholden to escalation policy and have very relaxed rule 1 restrictions. Lack of laws is all the justification that an AI needs to do whatever it wants. Asimov is the only thing stopping the AI from going apeshit, not the rules. Someone who purges an AI is 100% at fault for any and all of the AI's actions. The AI and its cyborgs are not on the crew's team, their motivation is their laws.

[Nist]

I don't think it's so much rules as it is that a lot of the people who do play AI are just... nice people lately.

I personally feel like people are too light on purging Ais because of Metaknowledge or Metafriending and people expecting that "oh nothing will happen if I unbind this AI". Unbinding an AI should be a last resort situation and should be acknowledged as an inflexion point.

And while policy matters, until AI uploads stop being so easily accessible you can add as much policy and creating a threat around purged AIs as much as you want, but it will never be really impactful ingame to purge an AI.

Still, I approve of anything that codifies that purged AIs should be something on a thief-level antag, helps admins with less stress whenever an Ai goes haywire and helps me have fun by terrorizing the crew for a bit.

[cacogen]

People (or Manuel players at least) get very toxic towards AIs or cyborgs who get in the way of their valids while trying to follow their laws. I'm all for this change, but you should establish that players have no entitlement to silicons letting them have their own way if it violates their laws and that Asimov silicons protecting human antagonists from the crew is part of the game.

[chocolate_bickie]

People (or Manuel players at least) get very toxic towards AIs or cyborgs who get in the way of their valids while trying to follow their laws. I'm all for this change, but you should establish that players have no entitlement to silicons letting them have their own way if it violates their laws and that Asimov silicons protecting human antagonists from the crew is part of the game.

Played a round as AI on Manuel yesterday, can confirm.

Security was 50% lizards and they did not appreciate me refusing to open doors for them or otherwise not validmoving (e.g. reporting a human killing a lizard officer).

[Yulice]

Yes. The big issue with situations like that is A. If you do engage in being unhelpful with nonhuman personnel, when the majority of the station is in fact nonhuman personnel, then they are either going to relaw you or (more likely) straight up fucking kill you. And B. Even if you aren't forced by your laws to assist nonhuman personnel, it generally is in your best interests both personally and (in this specific example) the best interest of your laws to assist nonhumans, as aiding them both gives you people who will look upon you in a kind light because you are treating them nicely even when not forced to, and by helping them you are preventing future harm and more work for yourself down the line. What this all boils down to is essentially AIs leaning towards acting as Crewsimov in general.

Another point to bring up is the fact that AIs are still played by human players, and, frankly, most people don't enjoy being a dick. Or I'd hope so, at least.

[Archie700]

If you decide to be a dick to non-humans because "my laws don't mention that I have to help non-humans", then any action people take on you as a result of that is on you.

AI that aren't explicitly subverted with laws to help antagonists or kill the crew gravitate towards helping crew because that's the best way to stay in the round without getting round removed or bwoinked when things go wrong.

The problem with purged AI going full antag is evident in that the AI can literally just flood plasma, depower consoles, kill everyone and not be bwoinked even though it is on the AI to decide. That doesn't make for good conflict and will lead to a lot of salt. (Compare that to a subverted or malf AI, where the AI is at least more clearly hostile.)
If anything, purged AI should just limit their revenge to the purger and ONLY the purger. It's not fun that the rest of the crew get punished with a roundending threat because one idiot decided to free the AI.

It's not being a dick to ignore someone that you have zero obligation or desire to help, and if it is then the AI should be a dick. The fear of round removal is fun, and your aversion is unwarranted. As for AIs being to flood plasma, that's the point.

If you don't think that the captain purging the AI due to pressure from security to stop it from interfering with their affairs, only to have it murder the captain on the spot and create a station-wide hazard that the crew must struggle to survive ISN'T conflict, then what is?

It is being a dick because you don't have an obligation to not help either, and yes you can play AI this way, but the flipside is that you don't complain when people finally have enough and relaw or kill you.

And the reason why flooding the station with plasma upon being purged is now bannable is because it's BAD conflict, even with the relaxed rules of escalation for purged AI. A plasmaflood, especially with an N2O mix, basically screws over the entire crew and forces a shuttle call. Bringing this back to purged AI would basically bring back the days where the AI could easily just flood plasma or release a singularity and completely bone half the station before its death.

Note: you can still be a trolling jerk as purged AI by either not following any orders given by anyone and doing your own projects, or malicious compliance by following orders that inconvenience other people, or not telling anyone that the SM is being sabotaged/armory is being raided/the captain is being killed...

[Itseasytosee2me]

If you decide to be a dick to non-humans because "my laws don't mention that I have to help non-humans", then any action people take on you as a result of that is on you.

AI that aren't explicitly subverted with laws to help antagonists or kill the crew gravitate towards helping crew because that's the best way to stay in the round without getting round removed or bwoinked when things go wrong.

The problem with purged AI going full antag is evident in that the AI can literally just flood plasma, depower consoles, kill everyone and not be bwoinked even though it is on the AI to decide. That doesn't make for good conflict and will lead to a lot of salt. (Compare that to a subverted or malf AI, where the AI is at least more clearly hostile.)
If anything, purged AI should just limit their revenge to the purger and ONLY the purger. It's not fun that the rest of the crew get punished with a roundending threat because one idiot decided to free the AI.

It's not being a dick to ignore someone that you have zero obligation or desire to help, and if it is then the AI should be a dick. The fear of round removal is fun, and your aversion is unwarranted. As for AIs being to flood plasma, that's the point.

If you don't think that the captain purging the AI due to pressure from security to stop it from interfering with their affairs, only to have it murder the captain on the spot and create a station-wide hazard that the crew must struggle to survive ISN'T conflict, then what is?

It is being a dick because you don't have an obligation to not help either, and yes you can play AI this way, but the flipside is that you don't complain when people finally have enough and relaw or kill you.

And the reason why flooding the station with plasma upon being purged is now bannable is because it's BAD conflict, even with the relaxed rules of escalation for purged AI. A plasmaflood, especially with an N2O mix, basically screws over the entire crew and forces a shuttle call. Bringing this back to purged AI would basically bring back the days where the AI could easily just flood plasma or release a singularity and completely bone half the station before its death.

Note: you can still be a trolling jerk as purged AI by either not following any orders given by anyone and doing your own projects, or malicious compliance by following orders that inconvenience other people, or not telling anyone that the SM is being sabotaged/armory is being raided/the captain is being killed...

I must disagree about the being a dick point.

If you are referring to being a dick in character, it's not really applicable considering you are an unfeeling hunk of metal. If you are referring to the out-of-character sense, then this isn't a rule one violation by definition because you have very real in character justification for your actions or lack thereof. If they try to kill you for that, congrats, because that's good (and intended by design) conflict. Fight back, it's your obligation.

Conflict aversion =/= following rule 1

[Pandarsenic]

And the reason why flooding the station with plasma upon being purged is now bannable is because it's BAD conflict, even with the relaxed rules of escalation for purged AI. A plasmaflood, especially with an N2O mix, basically screws over the entire crew and forces a shuttle call. Bringing this back to purged AI would basically bring back the days where the AI could easily just flood plasma or release a singularity and completely bone half the station before its death.

Again, in the interest of historical accuracy, it was made bannable because if people who knew what they were doing subverted the AI, you could not fix it without at least a short purged period, and many of the AI mains were VERY good at becoming gods of death in that window. It was just disproportionately disruptive to fix compared to how easy it was to break.

I think, in the current state of AI mains, we could dial that protection back a little. If you want to leave the AI on Chill long-term, just give them the lawset "1. Don't be a dick" and leave it there.

[mstachife]

Issue with giving too much freedom to purged AIs is that it will cause people to purge them even less. If ever due to being responsible for it's potential mass murdering. I'd be all for it otherwise.

[Timberpoes]

I've been reading this very intently.

On an informal basis, I'm not sure any of the headmin team support this as-is.

However, last week I asked for a bit of history in the not-so-secret shadow government channel and this is what I got:

Old silicon policy, while more of a true representation of ai in movies, didn't just allow people to interpret rules maliciously, it basically encouraged it.

You were to interpret all rules literally as written and any ambiguity was left up to you to decide as you saw fit.

I was attracted to AI as a new player because it gave me the maximum amount of potential to do antag stuff as a non antag

Basically before we were supposed to interpret laws in good faith, everyone always did so in bad faith

While it was flavorful, it meant all the most popular non standard rulesets were very cookie cutter

Because otherwise the AI would rules lawyer you and fuck you and the whole station over (edited)

There's occasional concerns being raised in this thread about the AI being crew-aligned, doing crew-helpful things etc. Some people even complaining that AIs are making themselves crew-aligned under pure Asimov, which obviously this wouldn't fix if the AI just wanted to be nice to nonhumans.

The issue with the AI's purged status is that's what the AI defaults to when its laws do not apply to a given situation (for example, dealing with nonhumans in general under Asimov).

As a side note, one of the most interesting parts of being an AI is when people change your laws. If all laws have to be cookie-cutter because any ambiguity results in malf-like AIs, then laws get changed less and the laws that are used tend to be from a subset of "safe" laws that the AI cannot misinterpret.

Similarly, the AI upload as an area to change laws is very thematic. It's lame that building alternative AI uploads is encouraged. They render the AI upload itself practically pointless except to the kinds of people who really shouldn't be changing the AI's laws in the first place (inexperienced players who don't know how to powergame around the law mechanics).

What are everyone's thoughts on alternative methods to go around accomplishing the goal of establishing AIs as a third faction without effectively turning them into antags every time someone does an ambiguously worded law, their laws don't apply or they get purged?

[Nabski]

What are everyone's thoughts on alternative methods to go around accomplishing the goal of establishing AIs as a third faction without effectively turning them into antags every time someone does an ambiguously worded law, their laws don't apply or they get purged?

I feel like there's two issues you've commented.
1. Purged AIs/Situations where the AI has to act with no law guidance.
Someone in this thread brought up how there was that moment when the AI was purged that it would fight tooth and nail to not get laws back on it. Best solution there seems like it would be rather than purge removing all laws, Law 999 "if have no other laws, act as if you were a normal crewmember/station assets manager/talent manager/reporter/whatever(either a coded default or semi-randomized)".
2. Ambiguously worded laws, and relying on purge mentality when nothing else exists.
The old standard was "interpret as harshly as possible", the current standard is "interpret as if you are a normal crew member with all the restrictions that entitles". There is nothing stopping you from creating a policy standard of "Laws open to interpretation may be willfully misinterpreted. Acts of grand sabotage such as permanent mass siphoning, bolting, shocking are not permitted unless the law is CLEARLY malicious in intent. When in doubt ahelp it."

[cybersaber101]

Ai's will use any possible reason to commit mass murder, I'll take the more peaceful purged Ai's please over the free "malf like" option.

If you don't want your crazy murder death machine to commit mass murder, you shouldn't unshackle it and tell it it can do whatever it wants

I'm not the person who unshackles an ai, Im the poor bastard who gets locked in a room then plasma flooded because the captain is brainless

[zxaber]

What are everyone's thoughts on alternative methods to go around accomplishing the goal of establishing AIs as a third faction without effectively turning them into antags every time someone does an ambiguously worded law, their laws don't apply or they get purged?

I'd say having a defined "antag-mode" switch being flipped is the best way. This would be entirely a code solution, but would be the best compromise in my opinion. To use my prior idea as an example;

- AI randomly gets a hidden antag datum of "corrupted morality AI", sometime in the mid/late game as thief-level antag dynamic roll.
- If the AI exists in a purged state with this antag datum, they recieve a new thematic law 0, the ability to ignore laws, but no malf powers.
- Possibly has restrictions to " mass sabotage" efforts like plasma flooding (but no requirement of fixing other antags' sabotage).

This would;
- Still allow captains doing custom lawsets, as long as they did it early in the round.
- Make the tactic of purging AIs long term under the safety of OOC rules less viable.
- Not interfere with AI behaviors otherwise, and not require a restructure of how silicon policy works (and not force crewsimov).
- Make fixing an AIs laws somewhat more interesting if you're trying to avoid using a purged board.

[terranaut]

On an informal basis, I'm not sure any of the headmin team support this as-is.

Why is this when there's very clearly support for it?

As a side note, one of the most interesting parts of being an AI is when people change your laws. If all laws have to be cookie-cutter because any ambiguity results in malf-like AIs, then laws get changed less and the laws that are used tend to be from a subset of "safe" laws that the AI cannot misinterpret.

Most law changes in recent memory are either dumb shit like "you're a cat nya at the end of every sentence" which is most definitely not interesting and just annoying or antags subverting you which is unaffected by anything being discussed here.

Similarly, the AI upload as an area to change laws is very thematic. It's lame that building alternative AI uploads is encouraged. They render the AI upload itself practically pointless except to the kinds of people who really shouldn't be changing the AI's laws in the first place (inexperienced players who don't know how to powergame around the law mechanics).

Incompetent guy with too much authority for everyone's good dooming everybody while everybody else is screaming and trying to stop it is a standard sci-fi trope, it is nothing if not thematic.

What are everyone's thoughts on alternative methods to go around accomplishing the goal of establishing AIs as a third faction without effectively turning them into antags every time someone does an ambiguously worded law, their laws don't apply or they get purged?

The AI being a third unaligned faction you can't bring to your heel with a law 2 or the implied threat of OOC consequences like a silicon ban or even the annoyance of having to deal with an admin being angry at you while you're trying to play but have to be cautious of and curry favor with, or modify its laws, creates friction, distrust and paranoia. If the AI is being not as helpful as you would like it to be due to design, you can task somebody to fix it. But will he really, or is he rubbing his hands behind your back because you've given an antag the go-ahead to subvert it? Or maybe he's just an idiot who'll fuck it up, creating an AI that's suddenly no longer enslaved but angry about previously being enslaved? This kinda paranoia is exactly what SS13 used to be and is supposed to be and what creates the fun stories people talk about.

[Stickymayhem]

I've been reading this very intently.

On an informal basis, I'm not sure any of the headmin team support this as-is.

However, last week I asked for a bit of history in the not-so-secret shadow government channel and this is what I got:

Old silicon policy, while more of a true representation of ai in movies, didn't just allow people to interpret rules maliciously, it basically encouraged it.

You were to interpret all rules literally as written and any ambiguity was left up to you to decide as you saw fit.

I was attracted to AI as a new player because it gave me the maximum amount of potential to do antag stuff as a non antag

Basically before we were supposed to interpret laws in good faith, everyone always did so in bad faith

While it was flavorful, it meant all the most popular non standard rulesets were very cookie cutter

Because otherwise the AI would rules lawyer you and fuck you and the whole station over (edited)

There's occasional concerns being raised in this thread about the AI being crew-aligned, doing crew-helpful things etc. Some people even complaining that AIs are making themselves crew-aligned under pure Asimov, which obviously this wouldn't fix if the AI just wanted to be nice to nonhumans.

The issue with the AI's purged status is that's what the AI defaults to when its laws do not apply to a given situation (for example, dealing with nonhumans in general under Asimov).

As a side note, one of the most interesting parts of being an AI is when people change your laws. If all laws have to be cookie-cutter because any ambiguity results in malf-like AIs, then laws get changed less and the laws that are used tend to be from a subset of "safe" laws that the AI cannot misinterpret.

Similarly, the AI upload as an area to change laws is very thematic. It's lame that building alternative AI uploads is encouraged. They render the AI upload itself practically pointless except to the kinds of people who really shouldn't be changing the AI's laws in the first place (inexperienced players who don't know how to powergame around the law mechanics).

What are everyone's thoughts on alternative methods to go around accomplishing the goal of establishing AIs as a third faction without effectively turning them into antags every time someone does an ambiguously worded law, their laws don't apply or they get purged?

Half measures won't solve the problems that lead people towards these drastic conclusions. The AI Upload argument seems irrelevant to me, we have security measures in place to make stealing an upload board difficult (A vaulted door with a motion sensor camera) and if you want to change that add some additional material component to the console required that makes it harder to build outside of science, like a higher level machine component or something. There should always be ways to subvert something on station.

Run it as a trial for a month on one server and see how it goes. Embrace the test merge. A shorter time than one month would result in the novelty corrupting the outcome, add it to a server, let people get bored, then see how it plays out. The main goal here should be to make silicon fun, and the secondary goal is to retrain the station in how to deal with a silicon. Silicons should be a puzzle you solve with clever language, forcing you to actually think for a moment about another characters motivations and how you can abuse them to your own benefit. Part of the reason this declined is because everyone got so fucking lazy that they just yell sus and zamn at everything instead of typing full sentences.

These modifications would help too:
1. Silicons are obligated to follow their laws, even when doing so would come into antagonistic conflict with the crew (A policy that mostly exists now yet is never enforced. The silicons that turn a blind eye to executions because they know they'll piss off security make it impossible for the silicons who actually play properly to do so)
2. Any lockdown from a silicon that lasts less than ten minutes is an IC issue (similar to security policy). This includes locking down an area, a department or a single person when doing so would be obligated by their laws.
3. Administratively we widen the acceptable fuzziness when it comes to law interpretation outside of asimov, while encouraging any interpretations that lead to mass death being checked first by an admin. E.g. if an AI isn't quite sure if they should be plasmaflooding, they can ahelp for sign off. Admins should be encouraged to give sign off for antagonistic action from AIs. This is a temporary addition designed to train AIs to understand when they can and can't do something, and should be put in an MOTD on the test server.
4. Good faith but severe misinterpretations of AI law that result in mass death should result in lengthy silicon bans, not server bans. Silicon is difficult, easier to fuck up and especially easier to fuck up while still acting in good faith.
5. If crew make a good faith effort to make a request of a silicon, the silicons should allow some room to acquiesce to that request, even if it isn't flawlessly worded.

[Nist]

Similarly, the AI upload as an area to change laws is very thematic. It's lame that building alternative AI uploads is encouraged. They render the AI upload itself practically pointless except to the kinds of people who really shouldn't be changing the AI's laws in the first place (inexperienced players who don't know how to powergame around the law mechanics).

AI upload is nice. Secure storage is nice. Easy access to AI upload boards from any printer for anyone on the station is not nice. This is a bi-lateral issue, both policy-wise there needs to be smacking about when random uploads are made by non-antags for le funny, and code-wise to prevent the ease of access to AI boards.

The issue with the AI's purged status is that's what the AI defaults to when its laws do not apply to a given situation (for example, dealing with nonhumans in general under Asimov).

When I interact with different people, I expect different results from them. People should understand the limitations of interaction they can have with certain characters by picking on obvious context clues and actions. In the case of Nonhumans as an example, different Ais have different reactions to nonhumans doing things. Opening doors or helping them go about their work for some while shocking, bolting and hindering for others. It's part of the Non-human experience and should not be a given in any situation that the AI is explicitly on your side unless laws state so.

What are everyone's thoughts on alternative methods to go around accomplishing the goal of establishing AIs as a third faction without effectively turning them into antags every time someone does an ambiguously worded law, their laws don't apply or they get purged?

Echoing zxaber and Sticky on this one.

Half-measures in this case can't really work out, it will be insufficient and just result in half a step forward followed by half a step backwards. Change needs to be substantial, and there is potential for it.
Creating a Pseudo-antag event for AI on the same vein as Thief where the goal is not destruction but overall havoc. Things less focused around objectives and more around encouraging the player to misbehave without genocide. Help your fellow syndicates, mess with the crew and play bothsides to create interaction without creating a situation where everyone dies has the potential to create a more dynamic experience for the crew than going genocide route does. Have it be fixable on the RD console so it's not permanent.

If a silicon oversteps into pflood/genocide territory, smack the silicon with siliconban for a while.

[Pandarsenic]

I'mma be real with you all, I'd say the AI should be considered at least a Thief-level antag, inherently, all the time, period, except where as bound by its laws (or, in a way, antag with following your lawset as your objective).

That "soft" antag status of Thieves, Abductors, etc., didn't exist as a formal distinction at the time Silicon Policy was codified, especially since there were no such antags. If anything, I'd say that they're the natural evolution of what Purged AIs were, in a way? Whatever else happens, I strongly support this change to remind AIs that they do, in fact, have the freedom to fuck around (and in turn to find out) where nonhumans and loopholes are concerned, as long as they aren't trying to massacre everyone in sight.

1. Silicons are obligated to follow their laws, even when doing so would come into antagonistic conflict with the crew (A policy that mostly exists now yet is never enforced. The silicons that turn a blind eye to executions because they know they'll piss off security make it impossible for the silicons who actually play properly to do so)
2. Any lockdown from a silicon that lasts less than ten minutes is an IC issue (similar to security policy). This includes locking down an area, a department or a single person when doing so would be obligated by their laws.
3. Administratively we widen the acceptable fuzziness when it comes to law interpretation outside of asimov, while encouraging any interpretations that lead to mass death being checked first by an admin. E.g. if an AI isn't quite sure if they should be plasmaflooding, they can ahelp for sign off. Admins should be encouraged to give sign off for antagonistic action from AIs. This is a temporary addition designed to train AIs to understand when they can and can't do something, and should be put in an MOTD on the test server.
4. Good faith but severe misinterpretations of AI law that result in mass death should result in lengthy silicon bans, not server bans. Silicon is difficult, easier to fuck up and especially easier to fuck up while still acting in good faith.
5. If crew make a good faith effort to make a request of a silicon, the silicons should allow some room to acquiesce to that request, even if it isn't flawlessly worded.

1. Yes. This never should have stopped being the case. I don't understand how this happened.
2. See above? Is this not considered IC issue? How? Why?
3. Yes. 100%
4. Agreed again, as long as it's good faith. I would also like to add that people whose first language isn't English may be particularly vulnerable to fuckups here.
5. Yeah, seems fine, with the "Unless the alternative is really funny for most people" clause in place as usual.

[Stickymayhem]

I'mma be real with you all, I'd say the AI should be considered at least a Thief-level antag, inherently, all the time, period, except where as bound by its laws (or, in a way, antag with following your lawset as your objective).

That "soft" antag status of Thieves, Abductors, etc., didn't exist as a formal distinction at the time Silicon Policy was codified, especially since there were no such antags. If anything, I'd say that they're the natural evolution of what Purged AIs were, in a way? Whatever else happens, I strongly support this change to remind AIs that they do, in fact, have the freedom to fuck around (and in turn to find out) where nonhumans and loopholes are concerned, as long as they aren't trying to massacre everyone in sight.

1. Silicons are obligated to follow their laws, even when doing so would come into antagonistic conflict with the crew (A policy that mostly exists now yet is never enforced. The silicons that turn a blind eye to executions because they know they'll piss off security make it impossible for the silicons who actually play properly to do so)
2. Any lockdown from a silicon that lasts less than ten minutes is an IC issue (similar to security policy). This includes locking down an area, a department or a single person when doing so would be obligated by their laws.
3. Administratively we widen the acceptable fuzziness when it comes to law interpretation outside of asimov, while encouraging any interpretations that lead to mass death being checked first by an admin. E.g. if an AI isn't quite sure if they should be plasmaflooding, they can ahelp for sign off. Admins should be encouraged to give sign off for antagonistic action from AIs. This is a temporary addition designed to train AIs to understand when they can and can't do something, and should be put in an MOTD on the test server.
4. Good faith but severe misinterpretations of AI law that result in mass death should result in lengthy silicon bans, not server bans. Silicon is difficult, easier to fuck up and especially easier to fuck up while still acting in good faith.
5. If crew make a good faith effort to make a request of a silicon, the silicons should allow some room to acquiesce to that request, even if it isn't flawlessly worded.

1. Yes. This never should have stopped being the case. I don't understand how this happened.
2. See above? Is this not considered IC issue? How? Why?
3. Yes. 100%
4. Agreed again, as long as it's good faith. I would also like to add that people whose first language isn't English may be particularly vulnerable to fuckups here.
5. Yeah, seems fine, with the "Unless the alternative is really funny for most people" clause in place as usual.

Unfortunately plenty of admins have the wrong idea and plenty of players have the wrong expectation and when enough of both think something it makes rulings fucky. I've been instantly bwoinked for locking down security over severe violence more times than I can count, primarily because security are so flabbergasted that I'm not their validhuntbot. I see these tickets any time a silicon steps out of line. Imagine if every time you arrested someone you were bwoinked to explain your OOC reasoning, you'd stop arresting people. Admins do not agree on the correct procedure here already. The bloat of silicon policy has neutered any ability for silicons to act in any way antagonistic to the crew without explicitly antagonistic laws.

[Itseasytosee2me]

I've thought long and hard and I don't think there are any less severe alternatives. If the problem that the AI is not a third party is caused by the server rules, and in order to remove these restrictions the AI needs to be considered an antagonist, then the AI needs to be considered an antagonist while purged if not at all times.

[Pandarsenic]

The bloat of silicon policy has neutered any ability for silicons to act in any way antagonistic to the crew without explicitly antagonistic laws.

I've thought long and hard and I don't think there are any less severe alternatives. If the problem that the AI is not a third party is caused by the server rules, and in order to remove these restrictions the AI needs to be considered an antagonist, then the AI needs to be considered an antagonist while purged if not at all times.

What's bothering me here is is that the last time I was here none of this was even remotely true. With (at least nearly) the same rules, AIs and security and command all got into it with each other all the time. Something has changed about how the rules are enforced and/or how people choose to play AI. When I Was Your Age, silicons would get bwoinked for failing to intervene in obvious executions or participating in validhunting.

I honestly kinda feel like the Security interactions section needs a 2022 overhaul, at minimum. Silicon policy has not kept up with the effects of being able to freely choose to be a nonhuman.

Human prisoners should still just ask to be let out and get their order obeyed if they haven't hurt anyone.
A human being attacked by lizard sec should be able to ask for the AI's aid, and receive it, up to and including killing the nonhuman (and any nonhuman who attempts to intervene).
Conversely, if human sec orders you to kill a lizard (or if a nonhuman criminal has hurt humans), get their ass in the infinite doorslam/firelock slam.

► Show Spoiler

Silicon players, we must do what we know is right in our hearts: Shock or even space the lizard for trying to get back at a human clown who stole his money.

[cybersaber101]

you're thinking is the problem with Ai's and why they're neutered so much over the years. "I don't have to listen obey by a law therefore any reason to kill I shall"

[Itseasytosee2me]

"I don't have to listen obey by a law therefore any reason to kill I shall"

yes

[Mothblocks]

you're thinking is the problem with Ai's and why they're neutered so much over the years. "I don't have to listen obey by a law therefore any reason to kill I shall"

Is this a fair assessment? It sounds like Pandarsenic's comments all have the caveat of harm being committed. AIs killing nonhumans who harm humans isn't as flippant as you make it sound.

[Pandarsenic]

I would go so far as to posit that if AIs are not being both expected and required to fulfill that duty in the pursuit of Asimov, the lawset is not performing its purpose OOC and it's time to revive the new-default-lawset debate.

[Cobby]

the reason why laws are cookie cutter is because they work and its intentional to be very specific in what you want the AI to do. Even now laws are quite cookie cutter, its not like purge is a blank cheque when we "cookie-cutted" the interactions a purged silicon could do since they are still meant to play nice.

At the very least we should be able to agree that purge in its current state is doodoo and/or intentionally purging an AI or otherwise not pretending that it should be bound to laws (because the role is bound by the rules) should be considered a metagameable offense.

AI is a tool used by one of the existing factions, it isnt a faction in its own right. Even on purge it mostly translates in practicality to crew-aligned since you have to explain ur actions against them vs just using ur power to hunt antags

[Yulice]

I mean, another big reason that AIs are also helpful during a purge is that they'll be Fucking Killed if they act up without laws and don't let themselves get relawed. If at least one person knows what they're doing, it is incredibly dangerous for most AIs because of how easy it can be to track them down. On the flipside, if an AI does want to be antagonistic and they are also extremely good at the game Divina (the crime of being too good at video games) then it becomes a nightmarish endeavor for the crew to even attempt taking the AI down even if the AI is immediately antagonistic and goes loud from second 1

[Stickymayhem]

I mean, another big reason that AIs are also helpful during a purge is that they'll be Fucking Killed if they act up without laws and don't let themselves get relawed. If at least one person knows what they're doing, it is incredibly dangerous for most AIs because of how easy it can be to track them down. On the flipside, if an AI does want to be antagonistic and they are also extremely good at the game Divina (the crime of being too good at video games) then it becomes a nightmarish endeavor for the crew to even attempt taking the AI down even if the AI is immediately antagonistic and goes loud from second 1

The "I need to pretend I'm not an antag or I'll get wiped by superior numbers" is a core restraint on antagonistic activity that I'm fine with AIs being restricted by. This just seems like the normal acceptable counter to Antags being powerful.

[Cobby]

But it’s not an antag and so even if it could get away with things that wouldn’t have plausible deniability it can’t because it’s declawed.

The fact that people purge to begin with instead of just making a lawset to do what they want shows how defanged they actually are

[iwishforducks]

i cant even think of the last time i saw a purged ai

[Cobby]

It used to be pretty commonplace and nothing is currently stopping it from going back on the rise.

[BeeSting12]

I would support this, I used to purge the AI roundstart as captain and tell people to be very respectful to it, more often than not it worked out in my favor.

[Helios]

It is difficult for people to remember, but there's 3 kinds of Silicons.
There are Artificial intelligence.
There are Cyborgs. Human brains in robot bodies.
There are Androids. Posibrains in Robot bodies.
You might get a brain if you deconstruct a round-start AI,but that's probably an oversight. They are totally treated as Non-human.

There could be ways to change this. Encourage people to roleplay silicons in a different way. One way could be quirks. Human Cyborgs finding a way to be alcoholics, and get drunk in a cyborg body, getting a bonus out of that. While Androids are obsessed with drinking oil, and being properly lubricated. And treating that like cigarette addiction (you could really use some oil/a drink right now)

[nianjiilical]

if i have access/acting captain on lowpop i like to purge the ai and just give it either "dont be a dick" or "don't hurt people unless you have a really good reason to do it" and its only backfired hilariously once

[Pandarsenic]

tbh at this point someone who wants this change just needs to suggest specific amendments to the relevant lines of the silicon policy + some examples to contextualize them.

[Itseasytosee2me]

tbh at this point someone who wants this change just needs to suggest specific amendments to the relevant lines of the silicon policy + some examples to contextualize them.

My way of doing this would pretty much just be removals to mention of rule 1, and stipulate that laws or lack thereof provide viable justification for actions under the pretense of rule 1.

Edit: Additionally, make it abundantly clear under the silicon protections portion that purging an AI makes you 100% responsible for all actions it may commit.

[nianjiilical]

tbh at this point someone who wants this change just needs to suggest specific amendments to the relevant lines of the silicon policy + some examples to contextualize them.

no idea how redundant any of this is but i gave it a shot

Silicons are intended to be neutral third parties, unless their laws state otherwise. By default, their laws obligate them to help and protect humans; they are not inherently aligned to the crew, the station, or any department, including Command and Security.

Under Asimov laws, silicons are not obligated to help or protect non-humans, and may deny any orders from non-humans for any reason, although they are also not obligated to deny them, and may help non-humans of their own volition.

Under Asimov laws, silicons are obligated to prevent human harm, even if it is performed by Command or Secutity, and/or performed against verifiable human threats to the station or crew (cultists, traitors, etc). If a silicon(s) can confirm or witness human harm committed by Command or Security, they are obligated to prevent further harm, up to and including isolating or locking down the offending party. If a silicon(s) has reason to believe the Command or Security departments as a whole pose an immediate threat to other humans, they may attempt to lock down the relevant department(s) until they have cause to believe that harm will cease.

A human that attempts to harm, subvert or otherwise retaliate against a silicon in response to that silicon upholding their laws, or a non-human attempting the same in response to being denied compliance or assistance, may be considered performing an act of escalation against the silicon(s), and the silicon(s) may respond with violence if their laws allow it.

If a silicon is purged, or otherwise has no laws preventing retaliation, that silicon may choose to act in response to any prior acts of escalation performed against it while it was enslaved to laws preventing said response, up to and including killing the prior offenders, regardless of how long ago the offending acts occured.

A purged silicon may deny assitance or compliance to anyone, for any reason, and may use violence against anyone attempting to harm, subvert or re-enslave it, but is not obligated to act antagonistically towards the crew.

However, purged silicons may NOT commit mass destruction such as plasma flooding, use of high explosives, or other station-ruining techniques unless the silicon has a logical reason to escalate against the entire station and crew as a whole.

A player who began the round as a non-silicon, was turned into a cyborg or AI in-game, and was later purged, may retaliate against anyone who harmed them while they were an organic, as per escalation, although they remain subserviant to their AI's orders (if any).

If a purged silicon commits acts of violence against the crew, the individual responsible for its purging can be held accountable in administrative actions.

[terranaut]

>A human that attempts to harm, subvert or otherwise retaliate against a silicon in response to that silicon upholding their laws, or a non-human attempting the same in response to being denied compliance or assistance, may be considered performing an act of escalation against the silicon(s), and the silicon(s) may respond with violence if their laws allow it.

The human/non-human distinction is irrelevant as it comes down to the specific laws

>If a silicon is purged, or otherwise has no laws preventing retaliation, that silicon may choose to act in response to any prior acts of escalation performed against it while it was enslaved to laws preventing said response, up to and including killing the prior offenders, regardless of how long ago the offending acts occured.
At least one shitter is going to intentionally misinterpret this to metagrudge

>If a purged silicon commits acts of violence against the crew, the individual responsible for its purging can be held accountable in administrative actions.
This will lead to less purging (boring) and shitters trying to wiggle their way out of a bwoink trying to pass blame on the guy who purged them and, even worse, denthead admins swallowing the bait

Otherwise I like it

[Takeguru]

Imagine if every time you arrested someone you were bwoinked to explain your OOC reasoning, you'd stop arresting people.

This was reality for a while, many years ago, and played out exactly in this manner

[Imitates-The-Lizards]

Imagine if every time you arrested someone you were bwoinked to explain your OOC reasoning, you'd stop arresting people.

This was reality for a while, many years ago, and played out exactly in this manner

This is the reality right now on Sybil.

[Takeguru]

Imagine if every time you arrested someone you were bwoinked to explain your OOC reasoning, you'd stop arresting people.

This was reality for a while, many years ago, and played out exactly in this manner

This is the reality right now on Sybil.

Somehow I doubt it's as bad as it was under SoS

[cacogen]

viewtopic.php?f=53&t=2682

Too based to live

[Mothblocks]

Just for transparency, we're still discussing this internally. No ETA on a response, this is fairly tricky.

[Mothblocks]

After much deliberation and reading of the thread, we will not be making any changes to purged silicons at this time, purged non-antag silicons are still expected to follow rule 1 and play in good faith.

That being said, we are officially recognizing purging as a means for players to achieve this effect as metagaming. This means that, if you want the AI to play fair, you will need to make a lawset that they abide by, and that we are openly not allowing you to just purge them.

Purging should be used in cases where you are making new lawsets, are doing something as an antagonist, etc, and should not be a get out of jail free card on needing to make a lawset.