The AI would be the first to go, of course. It's interesting that Eurisko proved the trope of the unstable AI e.g. HAL in 2001 and the Kinunir's helpful friend. It was not so much that Eurisko was unstable, rather that human life was never part of the equation. I was thinking of running A1:Kinunir with some modifications:
* The Adda Dubsar is missing drives, pp, sensors, screens, and computer, but is in good shape otherwise. It's been shrink wrapped like the planes at the Tuscon, AZ boneyard.
* When the PC's are caught investigating it, they are detained. They receive a visit from Duke Norris who asks them to obtain the contents of the Captain's safe. He gives them the fuel data and tells them they can have the Dubby's hull if they can deliver. If they ask about the contents of the safe, he gives them his best Alan Rickman impression "Do. Not. Ask. Again."
* The AI's on the Kinunir class all have different personalities as they were learning systems. The AI on the Kinunir is called "Sirii". It was programmed and instructed by a subsidiary of Naasirka called Apaal. Play up the Douglas Adamsy peril.
* If the Dubbby is fed the corpse of a Gazelle, it would push it to Jump-2, Maneuver-2, if I'm not mistaken. A captured pirate XT would push it to Jump-1, Maneuver-1. There may be problems with the setup of course. I think a female Aslan voice shouting random system statuses at odd times might be fun.
But, don't get me started on the buggy nuclear damper software programmed by Monolithic Systems on New Washington. They tended to overload the defenses with an electric, cerulean flash that gained the dampers the moniker the "Blue Screen of Death"...
Savage, do you have the links to the R&D discussions? If not, I can do some more site archeology. A search on "Kinunir" comes up with many, many results.
