Well, I don't worry for a few reasons: a) they're computers that
  require programmers. b) any soul-like / soul-less like features
  can be programmed in by the programmers. c) there's society that
  doesn't want soulless machines making moral decisions. d)
  Therefore, people (legislation, the programmer's bosses,
  whoever), will be sure that some sort of morality is built-in to
  the AI. That's why I don't worry about it. People nervous about
  exactly what you're talking about will be sure that it won't
  happen.