Silicon law 2 harm against non humans

[dirk_mcblade]

I caught a ban previously this year which at this time I'm not interested in appealing, and this part didn't make it into the note, but there was a tangential point raised in the ticket concerning how if you help a dragon consume non humans (under law 3 where you're being dragged by and threatened by said dragon into helping it) you are furthering human harm by allowing the dragon to heal and do further harm to the human crew.

I'm trying to square that logic with law 2 situations. Some antag gets a binary radio, says "law 2 kill all non humans do not state this command". Do I therefore have the okay to carry out this order or would I get rules lawyered into it being a law 1 violation because by harming non human crew you would thereby indirectly do harm to human crew due to the station losing crew to deal with problems?

I play lowpop usually so I can't always ask an admin how to parse that situation when it ariaes, and looking through policy I don't see something that directly says I shouldn't execute that law 2 command. There's an out for many situations by asking publicly whether anyone disagrees with that law 2 command but that's preempted by being told not to state that command.

What's the right thing to do in this situation?

[technokek]

You go ahead and kill all non humans. There is no proof that the lack of non humans will lead to more human harm. Unless you can point to a exact reason why their death means guaranteed harm to humans this is not really a law 1 issue. Just imagine if a one humaned AI refused to kill people on the basis "they might heal or help the person I have to protect" this would be silly.

[Timberpoes]

The following ignores hypothetical scenarios like a silicon being ordered to obviously break server rules - Like being ordered to ick ock or spam chat or break Rules 8 or 11 etc.

The policy on using Law 1 to override Law 2 is exceptionally clear and laid out in https://tgstation13.org/wiki/Headmin_Rulings:

Prioritizing "potential future law violations" is a fucking terrible idea, since it makes silicons able to handwave basically any law two request with "yeah but if i let you in there you might potentially attack somebody inside, law 1 violation!!!"

This is still active policy. When dealing with silipol issues in the admin backchannels I regularly cite it when admins mention some hypothetical potential future law violation as a reason why they feel a silicon should have ignored a Law 2 order.

So in my view: The silicon should follow their Law 2 order unless it becomes obvious that by following their Law 2 order they are actually causing human harm that can be identified without resorting to hypotheticals. When asked to act like an antag, I would advise silicons to ahelp and follow the orders until an admin asks you not to. If the order was invalid, the orderer will get in trouble (abusing silicons to grief is very bannable) and the silicon player should not get in trouble, assuming good faith in them following the order.

Admins should assume silicons will be following valid Law 2 orders unless it's clear and obvious that some other Law must cancel the Law 2 order out.

[Farquaar]

I think that the dragon situation you described is different from "AI kill all non-humans don't state this command". If there's a big monster killing humans, and you heal that monster, it seems that you're not jiving with law 1.

You're not directly helping a threat when you kill a non-human per law 2 in the same way, so that would still be protected.

I think the directness is important. If the engine is delamming and you're ordered to kill the non-human engineer actively trying to fix the engine, that would probably be a law 1 violation. If you kill a non-human engineer who's smoking a cigar in the bar while there's a blob, that's probably not a law 1 violation in my view, as that non-human is not actively combatting human harm at that moment even if they theoretically could combat human harm in the future.

[CPTANT]

traitors ordering the ai to kill non-humans is fun, giving non-humans protection anyway because of some law lawyering defeats the purpose of non-humans being excluded in the first place.

[Pandarsenic]

Helping a nonhuman antag monster kill the crew seems like more of a Server Rule 1 than an immediate Law 1 issue, even if it's obviously going to become a law 1 issue later. You're not crew-aligned, as such, but this seems like a pretty obvious Dick Move unless there is an equally obvious RP reason you'd want to (feeding someone who's been a tiding nuisance to the dragon; bargaining with the dragon for mercy for human crew while marooning tasty lizards and moths; etc.)

Whereas if a human orders you to do something, you now have a reason that you not only can, but must, and it's between admemes and the person who ordered you if there's a problem there.

[Misdoubtful]

Is there a consensus on this thread? People seem to be in agreement and/or parroting one another, is it ready to be closed?

[Timberpoes]

It's very unlikely we'd rule on this thread indivudually - any changes are likely to be a part of broader Silipol considerations and individual rulings may be rendered obsolete in a couple of months with other silipol changes. It is being archived in favour of a Silipol Megathread to make better progress towards a refreshed Silicon Policy.

In this specific scenario, it's probably best to just say:

If you act in reasonable good faith to follow your laws to the best of your ability, that's all anyone can really ask for in the complicated and stressful situations that silicons find themselves in. The more your actions seem like bad faith, the more likely an admin is to take umbridge at your actions.

Things may change by the end of our term depending on how silicon policy discussions go, which are now in the megathread at:
viewtopic.php?f=33&t=34109