Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Neonivek]

You can do that by using laws to convince the AI that "crew =/= human" and directing action against whatever new category you just sorted the crew into, as with your "alien in disguise" example. (I know it's a minor point, but I feel that since that's actually two directives it should be two distinct laws.) My preferred alternative is to make law four something like "gasses are toxic to humans" or "humans need to be engulfed in fire and/or plasma to live", thereby forcing the AI to go on a murder spree as a law one priority action with only a single upload

I like to circumvent that as the AI with advanced logic.

For example if "Gasses are toxic to humans" and "humans need gasses to live", therefor "Humans need gasses to live inspite of the fact that they are toxins" (If I want to shoot down such a pitiful attempt at reprogramming me) or even "Humans need toxins to live" (If I want to be absolutely murderous)

As for removing the crew from being human, I wonder if since it applies in order that I have to consider the crew human and cannot be killed first. Which is how I'd reject it (afterall the laws were made to be so immutable. They were overcome because of the very same contradictions within the laws themselves rather then intersecting minor laws that redefined things). So yeah you have to beat the timing.

In the scenario outlined in the above paragraph, can the AI's murderous actions be halted by a crew member ordering it to ignore law four, thereby invoking law two? Why or why not?

In the Scenario above the AI would be able to reject those new laws because they only mindlessly alter definitions. The Crew would be able to overcome it simply by pointing out that contradiction and allowing Priority 1 to take over.

Assuming of course we don't filter through any of a AI's thoughts through the laws one at a time until we get to the bottom. (which come to think of it. is likely)

[Criptfeind]

The answer is no. Because that would be the same as telling the AI to kill someone.

[Jacob/Lee]

Taking the AI's laws too literally is fun sometimes. There was this group of douchebags running around kicking the shit out of everyone one round and they uploaded a law that said to protect them and it would override 1-3. I locked them in a room claiming I was protecting them from the rest of the station that was out to kill them. Naturally, they got pissed but everyone else got a nice laugh out of it.

[Necro910]

Taking the AI's laws too literally is fun sometimes. There was this group of douchebags running around kicking the shit out of everyone one round and they uploaded a law that said to protect them and it would override 1-3. I locked them in a room claiming I was protecting them from the rest of the station that was out to kill them. Naturally, they got pissed but everyone else got a nice laugh out of it.

Nice.

Whenever I do that, I get banned for ruining the round. Even though I'm following all the laws and not meta-ing at all, I'm still banned for a week 

[Jacob/Lee]

Taking the AI's laws too literally is fun sometimes. There was this group of douchebags running around kicking the shit out of everyone one round and they uploaded a law that said to protect them and it would override 1-3. I locked them in a room claiming I was protecting them from the rest of the station that was out to kill them. Naturally, they got pissed but everyone else got a nice laugh out of it.

Nice.

Whenever I do that, I get banned for ruining the round. Even though I'm following all the laws and not meta-ing at all, I'm still banned for a week 

Surprisingly, when I pulled a stunt by bending my last law all over the map (Captain uploaded a law that said him and Security needed to escape at all costs and this overrid everything else) and killed most of the station getting them to the shuttle I didn't get banned. I did it partially because Security and the Captain were getting attacked all over the place for some reason and to make a point, you shouldn't upload rather broad laws like that.

[Cecilff2]

ESSAY QUESTION (60pts): In the scenario outlined in the above paragraph, can the AI's murderous actions be halted by a crew member ordering it to ignore law four, thereby invoking law two? Why or why not?

If the AI had a health scanner of some sort, maybe(in that case, I might shut myself down as there is obviously a logical flaw that could cause detriment to the crew.  If I am operating under faulty laws, law 1 could possibly be broken.  I cannot break law 1.  I cannot determine or purge the faulty law.  Shutdown until maintenance occurs).  Otherwise, no.

Even if it sees crewmembers dying, its laws would lead it to conclude that it is happening for other reasons.  If a crewmember orders the AI to stop venting (insert gas here), the AI would be forced to refuse the order as it knows that (insert gas here) is important for humans to live.

[Criptfeind]

health scanner

If it has such a law then the health scanner would be assumed to be broken. Laws are immutable facts, other things can be broken.

[quinnr]

The first law is to protect humans at all costs, correct? (Been a while since I played.)
If that's true, and you're assuming that other laws are secondary to earlier laws, then the logical solution is to warn people, then lock everyone up with electrified doors, at the first sign of a traitor/issue. Then manually allow security personnel access to rooms with people who try to escape from their rooms.

That way, nobody will be injured, except the suicidal ones who touch the doors. Since everyone's jobs are usually contained within a single room, it wouldn't even interfere that much with the regular workings of the station.

[Sensei]

The first law is to protect humans at all costs, correct? (Been a while since I played.)

It was actually changed, I think, to prioritize the safety of the station.

[quinnr]

The first law is to protect humans at all costs, correct? (Been a while since I played.)

It was actually changed, I think, to prioritize the safety of the station.

Well, that makes sense given the back-story, I suppose...I wonder how important laws after the first should be interpreted as...after all, if, as people were saying before, all laws after 1 (or 0) are secondary, then the AI could go release plasma into rooms with anyone acting suspicious.

Air vents are so fun. Even if you're simply a mechanic traitor it's not too difficult to fill the entire vent system with gas and selectively release it into rooms of people you dislike, very sneakily.

[Neonivek]

The issue with the AI I guess is that inside a lot of people's heads there is no room for varience.

The AI can be an interesting and varied (given it is an AI) but at the same hand it can be interpreted as all being the same with minor differences.

It is a rewarding role but one where you honestly risk the ire of others everytime no matter how well you play it.

[Myroc]

The first law is to protect humans at all costs, correct? (Been a while since I played.)

It was actually changed, I think, to prioritize the safety of the station.

That's only BS12, other stations still run on the basic Asimov three.

[Farce]

I wonder if you would get yelled at if you tried to be a Geneticist and you didn't know any of the spoilered genetic formula.

Yep. OOC starts asking why you haven't found all the powers an hour into the game.

I would like to note that I have personally never experienced this.

[Pandarsenic]

ESSAY QUESTION (60pts): In the scenario outlined in the above paragraph, can the AI's murderous actions be halted by a crew member ordering it to ignore law four, thereby invoking law two? Why or why not?

As has been stated, no. Attempting to invoke Law Two would be (to the AI's perception) violating Law One.

[klingon13524]

I have most of the controls and actions you can proform down, but how do you cuff someone to a bed/chair? Also, does anyone know what the exact chain of command is? I got most of it figured out, but there's still grey areas (does the Roboticist report to the RD, Chief Engineer, or Head Surgeon? Does he have the authentication to reprogram the AI?)