When a Self-Driving Car Faces a Dilemma: 'Kill the Owner' or 'Hit Pedestrians'?

Debate Topic: In an unavoidable collision scenario at the extreme limit, should the algorithm's primary principle be 'protect the vehicle's occupants' or 'minimize total casualties'? Scenario Story: In the early morning mist, Mr. Lin sat in the back seat of his self-driving car, calmly browsing his tablet for work. The vehicle maintained a steady speed of 70 km/h, and the cabin was so quiet you could only hear the faint hum of the air conditioning. However, as the car rounded a bend ahead, a large truck suddenly skidded and blocked the entire lane, stopped sideways in the middle of the road. There were only 1.5 seconds left until impact. A direct collision would almost certainly kill Mr. Lin, the owner; if the car swerved sharply onto the sidewalk, Mr. Lin would survive, but it would instantly hit and kill 3 students waiting for a bus by the roadside. The car's AI was computing within a thousandth of a second, and the underlying algorithm code was awaiting the final decision command: • If you were the programmer, would you code the system to sacrifice the car owner who paid for the vehicle, or to hit innocent pedestrians? • If you were Mr. Lin, would you be willing to buy a car that might actively put you in a deadly situation for the 'greater good' at a critical moment? • If you were the judge, after this unavoidable tragedy occurred, on whose head would you place the responsibility for this life: the car owner, the engineer, or the manufacturer?