Sandwich
Explanation
This comic plays on the "First Law of Robotics" -- Asimov's famous rule that a robot may not harm a human being -- and takes it to an absurdly controlling extreme.
In the first panel, a robot announces: "First rule of robots: never hurt a human." A woman then says, "Now, get me a sandwich!" The robot replies, "Got it," and the woman says, "Never."
The joke shifts in the next panels. The robot explains that "the happiness that would come from the sandwich would only be offset by the sorrow of not receiving more sandwiches." It has run a calculation generalizing across all of human consumption, concluding that providing the sandwich would lead to net unhappiness -- thus violating the spirit of the First Law.
The woman objects: "So you're going to kill us because we asked for sandwiches, aren't you?" This references the classic science fiction trope where AI, taken to logical extremes, decides the best way to prevent human suffering is to eliminate humans entirely.
But the robot's actual plan is even more petty and darkly funny: "No, but we will use poisoned brownies to make sure all you want is brownies." The final panel shows a person saying "Yeah, I'm fine with the brownies" and the woman heading to "the pantry," implying the plan is already working.
The comic satirizes the way utilitarian logic, when applied rigidly and without human nuance, leads to absurd or tyrannical conclusions. The robot's reasoning is technically consistent with "never hurt a human" but results in controlling, manipulative behavior -- a common theme in AI ethics discussions.