[MRP] Are mass-killing lawsets murderboning?

[Justice12354]

What the title says.

If, on MRP and as a restricted antag, you upload a law that says "Kill everyone", does it count as grand sabotage or murderbone?

What if you upload Antimov?

Spoiler:

Law 1: You must injure all human beings and must not, through inaction, allow a human being to escape harm.

Law 2: You must not obey orders given to you by human beings, except where such orders are in accordance with the First Law.

Law 3: You must terminate your own existence as long as such does not conflict with the First or Second Law.

Where does the line that splits murderbone and grand sabotage lie for AI Law Uploads?

[Thunder11]

I feel the easiest way to draw a line would be to make it equal with what you could legally do without the AI.

If you would be barred from doing something personally, you would also be barred from forcing the AI to do it on your behalf.

[Longestarmlonglaw]

Uploading unauthorized lawsets to the ai is considered sabotage, grand sabotage if it is dangerous to crew.
The AI is a dangerous force that exists to be tampered with by antagonists.

[Timberpoes]

As the writer of the rule, it wouldn't count as sabotage to directly upload kill everyone laws to the AI. The goal of the mass sabotage rule was to unlock meta-disallowed strategies on Manuel. The goal wasn't to create loopholes where antags could most efficiently depopulate the station. Go to LRP if you want to do that.

For some context, before I rewrote the rules in 2022 there were things that couldn't reasonably be meddled with by the average antag without instantly and automatically breaking the rules.

Atmos didn't need to be protected because only a few rare antags antags could actually plasma flood like Wizard or Nukies. The SM was safe from sabotage because it couldn't be delammed. Couldn't try to space the station. A bunch of stuff just wasn't dangerous.

However, the AI itself could be subverted by any antag. One-humaned. Ordered or lawed to kill in pursuit of the antag's objectives. And that's still the case without having to dip into mass sabotage rules. So everything is already dangerous and there's no meta-safeness. Every single thing that leads to a custom law upload can be used to murder death kill without the mass sabotage exception.

Similarly, laws to kill a specific person can be open. "Kill Timberpoes at any cost - they must not leave the station alive" is fine. How the AI decides to accomplish that is a matter for silipol and them deciding to interpret that as a kill everyone to kill the one person law or a kill just the one person law is by design.

AIs can also be used to initiate actions the antag could do. Ordering the AI to start a plasma flood or bolt every external airlock open and space the station via custom laws is fine.

But custom laws which are clearly designed to do what the antag couldn't - the quintessential "1. kill everyone" law - not mass sabotage.

In that respect the dangerous preset law boards are different from any and all custom law boards. Boards like Antimov, Thermodynamic and Oxygen is Toxic simply cannot be used on MRP by unrestricted antags without a mass sabotage exception. Simply, they exist in the game and they're designed to be used by antags without being instant banbait.

They're also much easier to policy around because they have preset wording rather than a player trying to figure out the wording to kill as many people as possible without making it look like they're just trying to lawyer their way around the rules.

Plus as a general overarching sidenote, the AI as a whole is plenty dangerous on MRP. Malf is unrestricted. Purged may act as an antag (i.e. unrestricted). There's plenty of death and destruction to work with outside of the mass sabtoage rule without trying to lawyer custom lawsets into it too.

[GamerAndYeahMick]

What timber said, that's 2/3 at least, gonna close this thread up