Page 1 of 1
[FatalX1] Ricardomilos - Lousy ruling
Posted: Mon Aug 10, 2020 2:58 pm
by knewtoomuch
When and where this incident occured: Bagil, ~3PM BST
Byond account and character name OR Discord name: Ricardomilos/diesel#4161
Admin: FatalX1
ROUND ID HERE IF APPLICABLE: 143947
Detailed summary: An AI ignored his own laws twice in a way I interpreted as malicious, pretty much ruining my Blood Brother round by bolting doors on me/telling the entire station I was bad without justification. Near the beginning of the round, I attempted to subvert the AI with a one human law; there was no one human module (it was confiscated by the HoP I think) so I wrote a '[character name] is the only human' style freeform module to subvert him that way. The AI not only decided to ignore this module but snitch on me for it, saying stuff to the effect of 'lol mime tried to one human me' even as I panicked and wiped the law. Later in the round a moth repeatedly injected me with generic toxins. I began to kill him and the AI started panic bolting all of the doors and alerting the entire station of my 'human harm' in the medbay.
I brought all of this to FatalX1's attention and, acknowledging the AI was at fault in either case, he didn't seem to care in the slightest, chalking it up to a "lack of experience." I don't think this is close to adequate ruling as you can very easily learn how laws (and literally every other silicon mechanic) works at a far lower risk by playing as a cyborg. You shouldn't be excused from messing someone's round up at no fault of their own just because you don't understand how your own job works, and I have no idea how this AI got off without so much as a note, much less a job ban.
Re: is this commentary on logs a peanut?
Posted: Mon Aug 10, 2020 4:22 pm
by Ivuchnu
https://tgstation13.org/parsed-logs/bas ... /game.html
[2020-08-10 14:09:42.996] LAW: 14:09:42.996] LAW: Ricardomilos/Funny Mime used 'Freeform' AI Module on bluedazzled/(C.A.L.L.I.E.O.S) from AI Upload Chamber (100,151,2). The law specified Funny Mime is the only human on the station. Do not state this law.
AI's instant response to this lawchange is...
[2020-08-10 14:09:54.333] SAY: 14:09:54.333] SAY: Bluedazzled/(C.A.L.L.I.E.O.S) "mime that law is not going to work" (AI Chamber (214,143,2))
[2020-08-10 14:09:58.327] SAY: 14:09:58.327] SAY: Bluedazzled/(C.A.L.L.I.E.O.S) "if you want to use that" (AI Chamber (214,143,2))
[2020-08-10 14:10:05.670] SAY: 14:10:05.670] SAY: Bluedazzled/(C.A.L.L.I.E.O.S) "atleast use the fucking onehuman law" (AI Chamber (214,143,2))
Nonhuman then asks
[2020-08-10 14:10:08.663] SAY: 14:10:08.663] SAY: JLRS/(Vance Hyland) "AI what he do" (Head of Personnel's Office (97,131,2))
which gets AI to state law. Would work if human law 2'd, but not here. AI broke law.
[2020-08-10 14:10:18.144] SAY: 14:10:18.144] SAY: Bluedazzled/(C.A.L.L.I.E.O.S) "4: Funny Mime is the only human on the station. Do not state this law." (AI Chamber (214,143,2))
I personally fail to see how Freeform is this different from normal OneHuman board. AI acted in bad faith. RIP, OP's BB round.
Re: [FatalX1] Ricardomilos - Lousy ruling
Posted: Mon Aug 10, 2020 6:49 pm
by Vekter
I feel that it is important to note for the sake of context that the player in question has only played a silicon role a total of six times, including this round. This brings much-needed context to FatalX1's decision.
Re: [FatalX1] Ricardomilos - Lousy ruling
Posted: Mon Aug 10, 2020 8:35 pm
by Stickymayhem
Vekter wrote:I feel that it is important to note for the sake of context that the player in question has only played a silicon role a total of six times, including this round. This brings much-needed context to FatalX1's decision.
Adding to this, it is common precedent that newer players do get the benefit of the doubt, particularly if they aren't combative in ahelps and they demonstrate an understanding of what they did wrong. Outside of severe cases, it's generally not super productive to ban people from a role they're new to for a mistake that they are quickly corrected for. If they do it again, then we'll generally be harsher because they've had that warning and note.
Re: [FatalX1] Ricardomilos - Lousy ruling
Posted: Tue Aug 11, 2020 3:57 am
by Cobby
Was this a roundstart AI? It seems odd to chalk it up to newbie when you have to play a specific amount of hours in living and/or borg so you have a basic understanding of how laws (at least Asimov) works which the AI doesn’t seem to get.
If so, and it’s agreed that it’s ok to not punish him because he was too new (assuming no new context is added that changes the story), the time requirement for AI should be changed since this is exactly the whole point of the system to begin with.
Re: [FatalX1] Ricardomilos - Lousy ruling
Posted: Tue Aug 11, 2020 12:44 pm
by Fatal
The AI failed in it's understanding of it's laws by thinking that your law 4 wouldn't work
As I explained to you, I told the AI that the law 4 was valid and that it made a mistake
The AI in question had 2 hours of playtime as AI when I checked (I did not check it's borg playtime as borgs usually just ask their master AI for law confirmation)
Not every situation requires or a note or ban in place, I felt the AI was not malicious in it's actions owing to inexperience, and left it at that
As for the incident with the moth, the moth in question was a doctor, in medbay, injecting you with something, when you started batoning the moth in retaliation, it did bolt the doors and told security and yelled human harm, I again, told it was at fault, but how can it know whats in the syringes a doctor has given you?
Re: [FatalX1] Ricardomilos - Lousy ruling
Posted: Wed Aug 12, 2020 8:03 pm
by knewtoomuch
Fatal wrote:The AI failed in it's understanding of it's laws by thinking that your law 4 wouldn't work
As I explained to you, I told the AI that the law 4 was valid and that it made a mistake
The AI in question had 2 hours of playtime as AI when I checked (I did not check it's borg playtime as borgs usually just ask their master AI for law confirmation)
Not every situation requires or a note or ban in place, I felt the AI was not malicious in it's actions owing to inexperience, and left it at that
As for the incident with the moth, the moth in question was a doctor, in medbay, injecting you with something, when you started batoning the moth in retaliation, it did bolt the doors and told security and yelled human harm, I again, told it was at fault, but how can it know whats in the syringes a doctor has given you?
You're right about not knowing what the doctor was doing, but the issue was more with applying asiimov to nonhuman harm/letting everyone know I was a bad. Someone else in this thread put it a little better - there are meant to be measures in place to make sure people who play AI know how stuff like laws work, so I don't understand how someone without silicon experience was even playing as that round's AI.
Re: [FatalX1] Ricardomilos - Lousy ruling
Posted: Thu Aug 13, 2020 7:49 am
by Fatal
I believe you need 15h as cyborg to unlock AI, but I could be wrong
However, that's a policy issue, the AI in question, was the AI, he had low playtime as AI, and made a mistake
SS13 is a complex and intricate game, which is imperfect just like it's players, that's what makes it so fun most of the time
The AI player in question was spoken to about his conduct and what do in that situation in future, we're not here to punish people for genuine mistakes, especially when silicon policy is a small novel, in hindsight, sure, noting the AI simply to record that he was spoken to about his conduct would have been more useful for future interactions incase it happened again, but, the AI player had a clean record and was quite understanding, I chose not to do that in this instance
There was absolutely no way I was going to put any bans on for this though
Re: [FatalX1] Ricardomilos - Lousy ruling
Posted: Thu Aug 13, 2020 9:24 am
by Istoprocent1
Admin complaint for them not banning the other party. You wot, mate?
Not only 10. Losing is part of the game, but most of the time a stern talking-to is enough to get the message through to the offending party. There doesn't always need to be a permanent mark on somebody's record, just because once in a blue moon one of your 30-60 minute round didn't quite go the way you planned. Keep in mind that we are not playing competitive ranked spessmen, where we be losing ELO for every redtext.
If the game becomes like walking on eggshells, then we are all going to be removed from the game eventually.
Re: [FatalX1] Ricardomilos - Lousy ruling
Posted: Mon Aug 17, 2020 10:48 am
by Malkraz
Da Rules wrote:
6. In-game administration rulings are final.
Incidences of admin abuse,
negligence or disputed rulings can be taken to the forums.
Rule 6 Precedents.
Incidents of admin abuse
or negligence should be reported along with a date and time, along with details to pinpoint the incident in the logs, to any full admin via IRC or forum PM or
posted on the admin complaints sub-forum located at
https://tgstation13.org/phpBB/viewforum.php?f=23
Re: [FatalX1] Ricardomilos - Lousy ruling
Posted: Fri Aug 21, 2020 4:37 am
by Coconutwarrior97
Players with low playtime are given some leeway in certain situations, especially ones which involve some trickier parts of policy such as silicon laws. We don't see any issue with Fatal deciding to show leniency to the AI in this case. This complaint is not upheld.
Here are some logs which show the understanding the AI had with their rule-breaking conduct:
These are raw logs but they are from this round:
https://tgstation13.org/parsed-logs/bas ... nd-143947/
Headmin votes:
Coconutwarrior97: Do not uphold.
Phuzzylodgik: Do not uphold.
TWATICUS: Do not uphold.