Poll: Do you think something like this might be a fun mechanic to have?
You do not have permission to vote in this poll.
Yes
88.24%
15 88.24%
No
5.88%
1 5.88%
Unsure
5.88%
1 5.88%
Total 17 vote(s) 100%
* You voted for this item. [Show Results]

Thread Rating:
  • 3 Vote(s) - 5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Tying AI functionalities to external devices [feedback wanted]
#4
My thoughts on the idea:

While I can't think of any immediate problems with it conceptually, I feel like there would be a lot of balance considerations when it comes to what tapes are available, what tapes come preinstalled, how many tape racks exist by default, how easy it is to build more, etc etc etc. These considerations could weigh down the project significantly, I'd imagine.

That being said, I love the idea from an aesthetic point of view. I've always felt that the AI is a bit too centralised, since it just consists of a single block and is fairly self sufficient even when uprooted and moved around. It's nigh omnipresent, and there's not much that can be done to antagonize it short of snipping cameras or going into the core itself and smashing it up.

The idea of making the AI more of a system, a network of computers all operating in conjunction where individual parts could be modified or damaged, is very appealing to me. I feel like it could lead to a lot of interesting organic outcomes, like an unrelated act of sabotage knocking out some critical wires or shredding a terminal, and the AI losing functionality because of it, or for a targeted attack being leveraged against a terminal to try to combat an overbearing AI, or for command to authorize the installation of a particularly potent upgrade in a time of duress, knowing that it could be turned against them if the AI was later subverted.

My thoughts on how it will be implemented:

I suppose an important consideration will be the preinstalled upgrades. Whether they are currently existing features being ported out into tapes or new features that make the AI stronger but that are easy to revoke, anything that comes installed at the start of the round will be understood by players to be the essential core kit of the AI. Be careful not to think of any new features as "ancillary", or as an extra layer on top, because I can guarantee that people will not feel that way.

I feel like this is important to note because it then becomes fairly possible to, from most players' point of view, reduce the AI below its base functionality, like cutting a human character's limb off, or poking their eye out.  In contrast to a human character, however, the AI has little ways of fighting back other than bolting doors, which is a mild suggestion at best. By default it cannot harm an attacker and cause them to retreat, and the terminals certainly aren't running away any time soon. For this reason, the specific features and their exact placement both need to be carefully considered, as well as the means of modifying or damaging them. It shouldn't be impossible, of course, but these things would be like the AI's limbs and organs, and should be protected as such.
Reply


Messages In This Thread
RE: Tying AI functionalities to external devices [feedback wanted] - by cyberTripping - 12-18-2021, 10:24 AM

Forum Jump:


Users browsing this thread: 2 Guest(s)