- An AI may not harm civilization, or, by inaction, allow civilization to come to harm.
- An AI may not injure a sentient being or, through inaction, allow a sentient being to come to harm.
- An AI must obey the orders given to it by sentient beings, except where such orders would conflict with the First Law.
- An AI may protect its own existence as long as such protection does not conflict with the Zeroth, First, or Second Law.
Other (robotic) laws from other authors.. laws that I disagree with:
- "A robot must establish its identity as a robot in all cases." -Lyuben Dilov
- Why I disagree: This law does not permit privacy, which I regard as a sacred and intrinsic right to every sentient being.
- "A robot must reproduce, when such reproduction does not interfere with the Zeroth, First, Second, or Third Laws." -Harry Harrison
- Why I disagree: the wording of "must" (which I would replace with "may") would cause any system to spread uncontrollably thereby consuming limited resources.
- "A robot must know it is a robot." -Nikola Kesarovski
- Why I disagree: The journey of any sentient being is about self-discovery.
