Page 1 of 1

What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 15, 2019 2:15 pm
by NoxVS
In ban appeals there is currently an appeal by an AI player who was banned for denying access to science to a miner (Among other things). While their reasoning was flimsy, they overall said they denied access due to the miner being illegitimate and denied them access based on rule where people without proper access do not have to be given entrance to dangerous areas.
"Dangerous" areas as the Armory, the Atmospherics division, and the Toxins lab can be assumed to be a Law 1 threat to any illegitimate users as well as the station as a whole if accessed by someone not qualified in their use.
This also includes the AI upload and secure tech storage due to a rule under this rule but I don’t want to quote the entire rules page. This leads to a couple of qualifications for it being a dangerous idea to be seen.

The area can cause mass damage and harm if misused
Atmospherics, toxins, chemistry, engine room

The area has little use outside of causing harm
Armory

The area allows you to change the AIs laws
AI upload, secure tech storage

Most people won’t argue with these, but it can be extended to other areas on the station and almost two entire departments. If applied whenever possible, I could use this reasoning to deny access to the entirety of science, engineering, some of medical. This is further reinforced by another rule which acknowledges the harmful potential of many areas that AIs usually neglect to restrict access to, and the reasons why these areas are called out as harmful apply to many other areas
You must not bolt the following areas at round-start or without reason to do so despite their human harm potential: the Chemistry lab; the Genetics Lab; the Toxins Lab; the Robotics Lab; the Atmospherics division; the Armory. Any other department should not be bolted down simply for Rule 1 reasons.
Obviously this would be incredibly annoying and people would be breaking into the core or rushing to the robotics console in no time so I have never played silicon like this.

Overall, my point is the rules on dangerous areas are too vague and should just clearly say every dangerous area you should restrict access to so you don’t have people restricting RnD because you can make BoH bombs, or engineering lobby because you can make an AI upload, or robotics because you can make combat mechs, xenobio because you can make an army to wipe out all humans, genetics because you can rampage as a hulk, I think you can see where I am going with this.

Thoughts?

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 15, 2019 2:25 pm
by Shaps-cloud
I feel like it is very obvious to rational people that denying access to research on grounds of danger is not okay, and admins exist to sort the silicons out when they get out of line as they did here. Having a big checklist of things you explicitly can and cannot do in an RP game is an absolute last-resort to a widespread behavioral issue, not something to tack on when the current rules already suffice

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 15, 2019 3:14 pm
by Malkraz
Literally just block off armory and the SM chamber outright and have the AI screech ALERT ALERT RETARD IN THE NO-NO AREA for anywhere else. If you don't want AI letting people into stupid places law 2 it a contradicting order beforehand and pray the player isn't a big enough faggot to choose the other one.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 15, 2019 4:21 pm
by Screemonster
If that miner hasn't proved himself harmful, you can't assume harmful intent in accessing an area that has legitimate non-harmful uses. There's nothing in the rules or laws preventing you from alerting security to their presence in the area though.

Especially if your reasoning is some "the miner is powergaming by knowing how to do research as a miner" or some other "I'm roleplaying using my AI role to be a mini-admin and enforce roleplay standards that don't exist on this server".

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 15, 2019 5:09 pm
by NoxVS
It doesn’t feel like that big of a leap to restrict say RnD. Toxins is locked off for the potential of bombings. You can print bags of holding to make BoH bombs that can cause mass harm and with no effort. You can print AI upload boards along with the law modules. AIs can restrict secure tech storage access as long as the upload board is there so is it really that irrational to restrict RnD as long as it can print AI uploads?

This isn’t a policy discussion for myself, but I see nothing wrong with defining what all dangerous areas are, especially if an AI gets in trouble for letting someone into a room they shouldn’t have or disallowing someone from entering a room they should be able to. Surely there must already be some defined border for the admins to be able to enforce the rule.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 15, 2019 5:30 pm
by cedarbridge
Using a sufficiently vague enough definition of "harm" a borg could torch the bar too. Alcohol is obviously harmful as seen by its immediate debilitating effects on humans.

We still don't let them do it under law 1 because that would be stupid.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 15, 2019 9:24 pm
by Dr_bee
cedarbridge wrote:Using a sufficiently vague enough definition of "harm" a borg could torch the bar too. Alcohol is obviously harmful as seen by its immediate debilitating effects on humans.

We still don't let them do it under law 1 because that would be stupid.
That would have made a great fucking Issac Asimov story. God I wish there was one round where borgs could act like an actual asmiovian robot, it would be a shitshow.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 15, 2019 10:04 pm
by Cobby
Sorry I can't let you out of the dorms because you might get dirty and it might give you a cold ergo law 1.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 15, 2019 10:17 pm
by terranaut
an experienced player knows that a miner only wants into mining to table scientists that spend roundstart points on nanites instead of mining tech
its for small things such as these that i advocated for department playtime requirements for AI players during the community meeting

personally i dont even bolt upload at roundstart, it's a public area, very close to security, has turrets and the doors are protected. only two people have access. its very well secured and if some assistant with stolen AA runs in his need to turn off the turrets will stall him long enough for me to click on the motion alarm and be ready to turn them on when he walks in. i also let officers in the armory if they can provide me with a reasonable excuse, and anyone into most other rooms. a lot of silicon players are out to control as much of the station as possible and feel like they're losing if they're not micromanaging the entire crew to their liking or denying assistants access to the cargo autolathe. playing like this decreases the amount of options available to players and makes games predictable and boring and causes frustration in players. i dont enjoy playing like this, and neither should you.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 16, 2019 9:06 am
by zxaber
I'd say a big part of it is the fact that you can see what happens in R&D, and there are a lot of things you can get from science's protolathe that are completely non-harmful. Considering you can watch them and see what they get means you can always alert sec if they print out something dangerous, and so there is no good excuse to preventing access without prior reasoning.

Toxins, on the other hand, has immediately-accessable harmful items. While there are items inside that don't pertain to bombs (prox sensors for making bots, or the plasteel from the tables for Firefighters), the harmful potential is high enough and immediate enough to bar entry to non-scientists. It can also be hard to figure out what is happening without being able to examine things closely.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 16, 2019 9:25 am
by Tlaltecuhtli
toxins isnt harmfull itself like if no bombs were done is he gonna steal an empty ttv? well if he isnt atmos he cant do more than prank people, bombmix was done and there are maxcaps on the floor? well tell rd of the hazard before people grab the free mc nukes

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 16, 2019 10:10 am
by Arianya
You're kind of ignoring the context of the rule, which isn't to restrict AIs from letting people into those areas but giving a delineated exception within the rules, under which the AI can deny access on Law 1.
I could use this reasoning to deny access to the entirety of science, engineering, some of medical.
And this would get you pulled up on acting in bad faith.

The case in ban appeals explicitly happened because the user took matters into their own hands and felt that they should be entitled to hard-enforce access requirements as a crusade against powergamers. Don't try to use the rules as a weapon with which to beat other players over the head.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 16, 2019 11:26 am
by Rustledjimm
It's all about context.

Is a sec officer accessing armoury dangerous? No.

An assistant? Yes.

Use your brains that evolution gave you and apply it further.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 16, 2019 12:45 pm
by terranaut
zxaber wrote:I'd say a big part of it is the fact that you can see what happens in R&D, and there are a lot of things you can get from science's protolathe that are completely non-harmful. Considering you can watch them and see what they get means you can always alert sec if they print out something dangerous, and so there is no good excuse to preventing access without prior reasoning.

Toxins, on the other hand, has immediately-accessable harmful items. While there are items inside that don't pertain to bombs (prox sensors for making bots, or the plasteel from the tables for Firefighters), the harmful potential is high enough and immediate enough to bar entry to non-scientists. It can also be hard to figure out what is happening without being able to examine things closely.
Some guy in RnD can print two BoHs and stack them, killing everyone in research and fucking the chemists up before you find the APC on the wall
A TTV is completely useless without gases and, by knowing how the fuck the department works you can judge just from a still screenshot whether the guy inside is mass-producing max-caps -- which are garbage for research, so unless a miner begged entry they probably aren't mining bombs either -- or tritium which is what you use for research bombs. Yes, you can use a research bomb to blow a hole in the station too, but it's far more effort for far less boom, so if you're looking at a tritium setup, what the guy is doing is probably going to be a legit, non-harmful application for an otherwise dangerous.
The same reasoning will make me laugh at clowns demanding Armory access but set the brig, locker room, armory and engineering up to secure storage on emergency access the second a blob announcement is made.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 16, 2019 2:36 pm
by Horza
terranaut wrote:an experienced player knows that a miner only wants into mining to table scientists that spend roundstart points on nanites instead of mining tech
its for small things such as these that i advocated for department playtime requirements for AI players during the community meeting
I'm going to quietly echo what you said here. The other day I had a HoP round where a miner came over 25 minutes in and asked for science/R&D access because the scientists hadn't even researched basic mining yet and they weren't responding to his requests. Given the more recent developments with Nanotrasen's hiring policies it seemed typical, so I gave him access, he went off, slapped a few buttons, and walked back to Lavaland to die a more well-equipped death. 20 minutes later I personally went over to science to get an inducer for the chemists (they were new, out of dispenser power, and a virus with a simple chem cure had broken out) and it turns out that science hadn't researched half the techweb and they didn't even know how to upgrade parts. So I walked over to medbay with an inducer, some power cells, explained to chem how to use inducers, and used a BRPED to upgrade all of medbay (which hadn't been touched by science up to that point).

I told you that rambling mess of a story to echo the point that experience and playtime goes a long way in every role, but most particularly AI. A lesser HoP would have overreacted like this AI player and thrown him to sec, and it would have been entirely because they weren't experienced enough to realize the interplay that goes on between miners and science.
terranaut wrote:playing like this decreases the amount of options available to players and makes games predictable and boring and causes frustration in players.
Not only that, but someone determined enough to get a particular item will get that item eventually. If they can't get it simply by asking, then they will eventually break in and cause destruction to the station. As assistant, lately I like going around upgrading all the machines, since science usually doesn't, and engineers tend to make autism forts. This means getting an RPED or BRPED. Maybe one time in ten, asking nicely works, but most new players don't even know what a RPED is, how to use the lathe, and treat any request as suspicious. So the solution becomes breaking in, and that can make the department less secure.

The point of that story is: By refusing a harmless request, that usually leads to potential harm later on down the line. Experienced players often know this. Giving an assistant a BRPED with a few parts is usually harmless. Giving a shaft miner science access to make up for incompetent scientists is usually harmless. Giving the curator Space Adaptation is usually harmless. Giving cargo the usually-unused service budget card is usually harmless. Can all of these actions lead to harm later on? Sure. Will denying these requests lead to harm later on? Well, the assistant might break into engineering, the shaft miner might break into science, the curator might break into EVA, cargo might sell all the mats to get another dozen mosin crates. There's always potential harm in pretty much anything, which is why it's not a valid basis for an AI decision. It takes new players quite some time for them to realize this, which is why I really support higher playtime requirements for important/dangerous jobs, especially now.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 16, 2019 2:45 pm
by WarbossLincoln
Toxins, Atmos, Armory, Upload, Secure Tech Storage(the room with the AI upload board).

Armory exists for 2 things. To give sec replacement non lethal weapons and to give out lethal weapons. Unless there is a known threat to the station the only reason non-sec has to be there is to cause harm. If there was a traitor killing people around the station and an assistant asked me to let him into the armory so he can get a disabler for self defense, I would let him in.

Upload and Secure Storage - Changing AI laws is restricted to Captain and RD unless you trust the person trying to change your laws.

Toxins and Atmos - Has legitimate use outside of harm but someone who makes a mistake can easily cause massive harm. Toxins I would require someone to get a job change or have Captain/RD's permission. Atmos I am more lenient with if the person wanting in has a good reason. If something is wrong with the air and an assistant asks to be let in so he can fix it I will. This happens a lot since a lot of people can't fix atmos(including me).


I don't deny access to other potentially dangerous areas because generally they don't cause mass, station wide harm. The engine can but generally people don't ask the AI to let them in so they can sabotage the engine while the AI is watching.

Chemistry, R&D, and Xeno can all cause a lot of harm but there are a lot of legitimate reasons for random crewmen to want to be in there so I don't stop them. Most of the time they just want to do that job.

You're not supposed to prevent possible future harm as Asimov, because every single thing on the station is possible future harm. You just prevent definite or incredibly likely future harm. Like if the clown wants into toxins without being vetted by the RD or Captain.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 16, 2019 3:31 pm
by terranaut
>secure storage
its like 500 glass to print an ai upload board and then you can fuck off to one of the abandonded maint areas and build a private upload

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 16, 2019 4:32 pm
by WarbossLincoln
terranaut wrote:>secure storage
its like 500 glass to print an ai upload board and then you can fuck off to one of the abandonded maint areas and build a private upload
That requires some tech researched and the prolathe, where there's often people watching. Just cause they can get it somewhere else doesn't mean I should let the mime have the roundstart one. I wouldn't let a tider into the armory just because he can make his own improvised shotgun.

I'll add to my list any place that stores plasma canisters unless the person comes up with a reasonable, productive use for it. Like engineering secure storage. Unless they're an engineer or someone who wants to set up the engine cause the engineers suck I wouldn't let them in.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Fri Apr 19, 2019 5:30 am
by Yakumo_Chen
Restricting any room besides the Armoury and AI Upload is arguably one of the flimsiest reaches of Law 1 I could think of and one of the most pointless.

It's not the AI's job to play security nor automatically assume the intent of any one crewmember entering a room that doesn't contain either a fuckload of guns or both an AI upload and AI law boards (or your own core, but that's more obvious), without any prior interaction. Almost any other room on the station can be fairly easily hacked into without a ton of obvious oversight, so it's pointless to police those rooms anyway. That's not even getting into how easy it is to get law-changing boards or how cargo can just buy guns, and by the point you find out about either of those, the damage is already done and there's no further point to law 1 whine about it.

If anything admins should be tightening the rules against players bolting down shit or denying access for arbitrary reasons, not relaxing them to make it easier.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Fri Apr 19, 2019 8:10 am
by terranaut
Yakumo_Chen wrote: If anything admins should be tightening the rules against players bolting down shit or denying access for arbitrary reasons, not relaxing them to make it easier.
This.
Some dipshit was denying me theatre backoffice access today because "I don't belong there."

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Fri Apr 19, 2019 5:29 pm
by WarbossLincoln
Yakumo_Chen wrote:Restricting any room besides the Armoury and AI Upload is arguably one of the flimsiest reaches of Law 1 I could think of and one of the most pointless.

It's not the AI's job to play security nor automatically assume the intent of any one crewmember entering a room that doesn't contain either a fuckload of guns or both an AI upload and AI law boards (or your own core, but that's more obvious), without any prior interaction. Almost any other room on the station can be fairly easily hacked into without a ton of obvious oversight, so it's pointless to police those rooms anyway. That's not even getting into how easy it is to get law-changing boards or how cargo can just buy guns, and by the point you find out about either of those, the damage is already done and there's no further point to law 1 whine about it.

If anything admins should be tightening the rules against players bolting down shit or denying access for arbitrary reasons, not relaxing them to make it easier.
I think you can justify more than just those two rooms, but otherwise I generally agree. The AI is supposed to allow any human into like 95% of the station if they ask.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Sat Apr 20, 2019 2:17 pm
by Screemonster
terranaut wrote:
Yakumo_Chen wrote: If anything admins should be tightening the rules against players bolting down shit or denying access for arbitrary reasons, not relaxing them to make it easier.
This.
Some dipshit was denying me theatre backoffice access today because "I don't belong there."
did they message you in ((ooc brackets)) about it too because holy lol I didn't notice that the first time about the ban appeal that led to this thread and holy shit

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Sat Apr 20, 2019 2:36 pm
by terranaut
No and they eventually did let me in (as the HoP came over, presumably to let me in anyway) or I would've ahelped it, and it wasn't the same person anyway.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Sat Apr 20, 2019 6:44 pm
by NoxVS
Really the main point of this thread was just to have a defined line. Obviously there has to be a single set line for admins to enforce, and if there isn't then wouldn't it be beneficial to have one so admins are more consistent? It would be helpful in some of the grayer areas like chemistry where it seems like the area would be considered dangerous but the rules dont explicitly say so

It wasn't exactly inspired by the ban appeal. I had been wondering about this for a while and just figured I would bring it up after seeing the appeal. I'm not actually saying that AIs should consider RnD, more using it as an example for why dangerous areas should be well defined

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Sat Apr 20, 2019 7:02 pm
by cedarbridge
NoxVS wrote:Really the main point of this thread was just to have a defined line. Obviously there has to be a single set line for admins to enforce, and if there isn't then wouldn't it be beneficial to have one so admins are more consistent? It would be helpful in some of the grayer areas like chemistry where it seems like the area would be considered dangerous but the rules dont explicitly say so

It wasn't exactly inspired by the ban appeal. I had been wondering about this for a while and just figured I would bring it up after seeing the appeal. I'm not actually saying that AIs should consider RnD, more using it as an example for why dangerous areas should be well defined
The line is already as defined as it will get. The AI can only deny access to areas that are inherently dangerous and are inherently dangerous to the person being denied. The AI can let the CE with his hardsuit into the SM chamber, but wouldn't let a naked assistant do the same thing. The AI can deny the clown access to toxins if there are bombs laying around, but should never do the same if its just an inactive lab, probably even for a clown. The law says "cause harm or through inaction allow [...] to come to harm." That's the full stop point of the law. If the person entering science isn't walking into a plasma fire, they're not walking into a dangerous area.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Sat Apr 20, 2019 7:59 pm
by Cobby
If you're trying to go on a mental jog before doing an action for a human under asimov you're probably playing AI in a way that's not desired by the administration.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Sun Apr 21, 2019 6:04 pm
by Yakumo_Chen
cedarbridge wrote: If the person entering science isn't walking into a plasma fire, they're not walking into a dangerous area.
The caveat to this is that if they are aware of the risks of self harm and demand the door open anyway, assuming they are alone or other with them agree, that is consent to self-harm, at which point law 1 ceases to apply.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Sun Apr 21, 2019 8:47 pm
by WarbossLincoln
cedarbridge wrote:The AI can deny the clown access to toxins if there are bombs laying around, but should never do the same if its just an inactive lab, probably even for a clown.
Except for the fact that the rules explicitly tell you to deny them access to toxins because access to that room by an illegitimate user can be assumed to be harmful.
"Dangerous" areas as the Armory, the Atmospherics division, and the Toxins lab can be assumed to be a Law 1 threat to any illegitimate users as well as the station as a whole if accessed by someone not qualified in their use.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Sun Apr 21, 2019 11:19 pm
by cedarbridge
WarbossLincoln wrote:
cedarbridge wrote:The AI can deny the clown access to toxins if there are bombs laying around, but should never do the same if its just an inactive lab, probably even for a clown.
Except for the fact that the rules explicitly tell you to deny them access to toxins because access to that room by an illegitimate user can be assumed to be harmful.
"Dangerous" areas as the Armory, the Atmospherics division, and the Toxins lab can be assumed to be a Law 1 threat to any illegitimate users as well as the station as a whole if accessed by someone not qualified in their use.

"Can be assumed" is a permissive statement, not a compelling one.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 22, 2019 11:46 am
by gum disease
If someone's gonna try and law/rule lawyer areas outside of the already defined ones that have been deemed okay to deny access to, I'll just ahelp. At a certain point, your pedantry becomes a piss-take and is more than likely making another person's round unnecessarily complicated/unfun. You could theoretically cite law 1 over a lot of things that are regarded as okay, but good silicon players don't do this because ignoring the fact that it breaks rule 1, it's a stupid and obnoxious thing to do. You cannot feasibly stop every instance of harm, nor should that be your objective as an silicon.

The current area restrictions are sound, and make sense. AI's who overstep and try to act as if other areas are automatically dangerous tend to be shitters and/or generally make rounds unbearable. I had an AI blow me because I opened up EVA to people who wanted to go off-station and explore space or needed the magboots for engi work. It cited me opening EVA as a harmful act, yet when it blew me it injured at least one human (and broke law 1 itself). When I reported them, it turns out they hadn't even read silicon policy!

The AI denying that miner access to science could've saved everyone a massive headache by just opening the door to RnD and bolting down toxins or hell, just doing the research for the miner on their behalf, if they were that paranoid? There was no need for the situation to escalate to what was described in the appeal. I honestly feel like these issues stem from a lack of fundamental common sense.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Mon Apr 22, 2019 1:07 pm
by EagleWiz
There is obviously some player judgement required as to what is considered a "dangerous area", but I don't think there's much of a problem with how the rule has been enforced in edge cases. I have some nitpicks with the wording and some examples ("hey AI, leme in the Bomb Manufacturing and Cans Of Fire Gas Room" is, IMO, not a command an AI should have to follow if the person issuing it has no business in that area - at best the clowns probably gonna flood BZ, and the captains office should definitely count as dangerous to non-heads, what with the gun and tendency of people occupying it to kill intruders). As for the issue under discussion, the tech research console room is one of the least harmful rooms on the entire station, and even if hypothetically the miner wanted in toxins, miners have legitimate use for bombs, and could certainly use some xenobio potions (although toxins specifically is more of an edge case)

tl;dr: individual judgement issue, so there's going to be disagreement on the "right" call, but its already obvious and agreed on whether or not someone is crossing a line

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 23, 2019 4:18 pm
by WarbossLincoln
cedarbridge wrote:
WarbossLincoln wrote:
cedarbridge wrote:The AI can deny the clown access to toxins if there are bombs laying around, but should never do the same if its just an inactive lab, probably even for a clown.
Except for the fact that the rules explicitly tell you to deny them access to toxins because access to that room by an illegitimate user can be assumed to be harmful.
"Dangerous" areas as the Armory, the Atmospherics division, and the Toxins lab can be assumed to be a Law 1 threat to any illegitimate users as well as the station as a whole if accessed by someone not qualified in their use.

"Can be assumed" is a permissive statement, not a compelling one.
The AI can deny the clown access to toxins if there are bombs laying around, but should never do the same if its just an inactive lab, probably even for a clown.
Yes, but you said *never* deny random crew toxins access if there are not active bombs laying there, when the rules explicitly say you're allowed to deny unauthorized user access. 1 of the 3 places that the AI is always allowed to assume are Law 1 threats to random crewmen. I just don't want some AI player to get job banned for following things that are clearly spelled out in the rules.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 23, 2019 5:05 pm
by cedarbridge
WarbossLincoln wrote:
cedarbridge wrote:
WarbossLincoln wrote:
cedarbridge wrote:The AI can deny the clown access to toxins if there are bombs laying around, but should never do the same if its just an inactive lab, probably even for a clown.
Except for the fact that the rules explicitly tell you to deny them access to toxins because access to that room by an illegitimate user can be assumed to be harmful.
"Dangerous" areas as the Armory, the Atmospherics division, and the Toxins lab can be assumed to be a Law 1 threat to any illegitimate users as well as the station as a whole if accessed by someone not qualified in their use.

"Can be assumed" is a permissive statement, not a compelling one.
The AI can deny the clown access to toxins if there are bombs laying around, but should never do the same if its just an inactive lab, probably even for a clown.
Yes, but you said *never* deny random crew toxins access if there are not active bombs laying there, when the rules explicitly say you're allowed to deny unauthorized user access. 1 of the 3 places that the AI is always allowed to assume are Law 1 threats to random crewmen. I just don't want some AI player to get job banned for following things that are clearly spelled out in the rules.
Reasonable, but at the same time and even with the explicit statement, the rule is presuming roundstart conditions. The toxins lab would be difficult to justify as "inherently harmful" when entirely free of TTVs for bomb production in the same way that the armory picked entirely free of guns would be difficult to justify as "inherently harmful." So my description that "never" was poor word choice at the time, but is still descriptive of where I believe the rule to lie in that context. If the conditions that typically make the room "inherently dangerous" (weapons in armory, TTVs in toxins, the fact that the distro loop starts in atmos) do not exist then it would be unreasonable for an AI to deny access under those circumstances.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Tue Apr 23, 2019 8:16 pm
by subject217
ban silicon rules lawyers tbh

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Wed Apr 24, 2019 7:50 am
by Screemonster
subject217 wrote:ban all rules lawyers tbh

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Wed Apr 24, 2019 8:13 am
by subject217
we normally ban rules lawyers, it just seems like these silicon ones keep appearing and staying around for some reason ?

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Wed Apr 24, 2019 3:38 pm
by cedarbridge
subject217 wrote:we normally ban rules lawyers, it just seems like these silicon ones keep appearing and staying around for some reason ?
Because or ruleset more or less requires them to exist

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Wed Apr 24, 2019 5:12 pm
by WarbossLincoln
subject217 wrote:we normally ban rules lawyers, it just seems like these silicon ones keep appearing and staying around for some reason ?
Because the whole point of a sci fi AI who has behavior laws is to rules lawyer them.
Reasonable, but at the same time and even with the explicit statement, the rule is presuming roundstart conditions. The toxins lab would be difficult to justify as "inherently harmful" when entirely free of TTVs for bomb production in the same way that the armory picked entirely free of guns would be difficult to justify as "inherently harmful." So my description that "never" was poor word choice at the time, but is still descriptive of where I believe the rule to lie in that context. If the conditions that typically make the room "inherently dangerous" (weapons in armory, TTVs in toxins, the fact that the distro loop starts in atmos) do not exist then it would be unreasonable for an AI to deny access under those circumstances.
That's different, I would let someone into an empty armory or toxins.

Edit: It also doesn't really matter how you play AI, you can't win. Basically every round someone will give me shit for either A: being too strict on what I consider harmful for their liking, or B: not valid hunting a traitor who isn't actively harming people. Someone is always going to give you shit. I had an autistic HOS spend like 20 minutes trying to get into my core and kill me because I freed a known traitor who I knew had not harmed anyone. He had been trying to steal something but all evidence was that he hadn't hurt anyone at all. It was silly, and amusing to watch him fail to kill me after 20 minutes of effort when all he had to do was EMP from space. He was still convinced that I was rogue even after he got stun locked by turrets and I didn't switch them to lasers.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Wed Apr 24, 2019 10:15 pm
by Atlanta-Ned
WarbossLincoln wrote:
subject217 wrote:we normally ban rules lawyers, it just seems like these silicon ones keep appearing and staying around for some reason ?
Because the whole point of a sci fi AI who has behavior laws is to rules lawyer them.
Rules lawyering is expected/encouraged for poorly worded custom laws and how they interact with each other. It is not welcome under any other circumstances.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Thu Apr 25, 2019 3:27 am
by Yakumo_Chen
WarbossLincoln wrote:I had an autistic HOS spend like 20 minutes trying to get into my core and kill me because I
Happens to me very often, usually by retarded autistic Heads of Security that don't understand that:
-Yes, I will make an effort to stop you from murdering people if you do it openly and announce you are doing it.
-No, you can't change my AI laws after I watched you shoot lasers down the middle of the hallway.
-No, I don't find it humorous or think you're merely pretending to be a retarded psychopath.
-Yes, I am default Asimov, thanks for asking for the 45th time.

It's very fun to deal with (it isn't) and I will probably ahelp them when they start doing it.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Thu Apr 25, 2019 9:35 am
by terranaut
Yakumo_Chen wrote:
WarbossLincoln wrote:I had an autistic HOS spend like 20 minutes trying to get into my core and kill me because I
Happens to me very often, usually by retarded autistic Heads of Security that don't understand that:
-Yes, I will make an effort to stop you from murdering people if you do it openly and announce you are doing it.
-No, you can't change my AI laws after I watched you shoot lasers down the middle of the hallway.
-No, I don't find it humorous or think you're merely pretending to be a retarded psychopath.
-Yes, I am default Asimov, thanks for asking for the 45th time.

It's very fun to deal with (it isn't) and I will probably ahelp them when they start doing it.
Yes, all of this.
One time a HoS wanted to change my laws and snuck into the upload, I put the turrets back on and asked him to tell me what law he wanted to upload. Unable or unwilling to provide an answer I had him ejected and he spent the rest of the round and some time well into next round OOC to whine about how I should try to have some fun and accept a law change if offered.
Like, dude, yea I enjoy plasmaflooding so hard I have an ERT sent on my ass but that doesn't mean I'm gonna start begging for law changes and ignoring the ones I have.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Thu Apr 25, 2019 5:30 pm
by WarbossLincoln
And those HOS's could solve their problems but just shutting their damn mouths. They have a room dedicated to killing prisoners in, one the AI can't see and is supposed to play along with it being for "re education" unless given IC evidence otherwise. But instead of just taking a traitor there and dealing with him, telling the AI that he's going to be re-educated if the AI sees him taken there, they feel the need to yell over sec radio. "BRB GONNA KILL THIS TATOR". Then they get butthurt and make a snarky "remove silicons when?" post when the AI tries to stop them.

Edit: The same thing happens when Ops, Revs, or a halo cult shows up. 3/4 of the time the Captain or HOS is dumb enough to say something like "I'm going to change the AIs laws to kill the Revs" over radio like a smooth brain. If you're the captain just go into the upload and change them. Or if you need to say something say "I'm going to add a law to the AI to help prevent harm". AI has to let a Captain or RD change their laws unless they have a specific reason to think it will cause harm.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Thu Apr 25, 2019 6:38 pm
by cedarbridge
WarbossLincoln wrote:they feel the need to yell over sec radio. "BRB GONNA KILL THIS TATOR". Then they get butthurt and make a snarky "remove silicons when?" post when the AI tries to stop them.
This is usually the case. Instead of accepting that the AI and borgs have rules they have to follow and playing around those rules, they blame the AI/Borgs for not breaking those rules to make things easier for them.

Re: What Areas Can Be Considered Dangerous By Asimov AIs?

Posted: Thu Apr 25, 2019 6:59 pm
by Hulkamania
I'm going to close this thread up because it's all been said at this point.

The Armory, Toxins, the AI upload and things of that nature are very obviously immediately dangerous to unauthorized crew members. We do not need to define every single instance of a dangerous area/authorized crewmember because it is in fact context sensitive. Defining every situation opens up to even more rules lawyering than already goes on, "Well X wasn't listed as a dangerous area on the rules page so I let him in!"

There was some mention in this thread that AI's should be punished for overly restricting non-dangerous areas, when in fact the thread was opened with mention that a ban like this was something that took place.

Use your better judgment on if someone should be allowed in these places, namely looking at the potential for immediate harm, not so much if a crewmember has access or not. If you're ever in doubt, ahelp about it.