AI experts suggest instilling human values ​​in robots to prevent betrayal

AsiaIndustrial NetNews: Beijing time on May 28, according to foreign media reports, with the development of artificial intelligence, more and more scientists are beginning to worry that artificial intelligence will bring various unpredictable effects to human beings, and may even be disasters. At the recently opened “TED2017” conference, computer expert Stewart Russell once again warned that things made by humans (artificial intelligence) may evolve to be smarter than humans.Russell suggested that humans mustRobotUncertainty is preset internally to guard against their betrayal.

AI experts suggest instilling human values ​​in robots to prevent betrayal

Stuart Russell, a professor of computer science at the University of California, Berkeley, explained that although he himself supports artificial intelligence research, it always feels that it is not necessarily a good thing to create a thing that is smarter than its own species. In order to prevent robots from taking over us, Russell suggests, we must make robots that are selfless, that is, there is some procedural uncertainty in their goal tasks, and that they should be taught about human values. By instilling “human common sense” into AI systems, scientists must redefine AI so we can build beneficial machines.

Russell said robots should be programmed according to certain laws or laws, such as the three laws of robotics proposed by science fiction writer Isaac Asimov, to prevent them from causing unintentional harm while performing their tasks. The main contents of the three laws of robotics proposed by Isaac-Asimov include: 1) Robots must not harm humans, or cause humans to be harmed by standing by; 2) On the premise of not violating the first law, robots must obey human beings. Command; 3) The Robot must protect itself without violating the first and second laws. Therefore, altruism, modesty, and other virtues of human values ​​must be programmed in when programming a Robot. Russell believes that “this is the first step for humans and artificial intelligence to adapt and live in harmony with each other.”

However, Tom Gruber, Apple’s senior R&D director for Siri, is more optimistic about the future of humans and artificial intelligence. He believes that artificial intelligence will be used to enhance human forgetting, so that humans will never forget anything. Gruber said, “It is inevitable that artificial intelligence will bring superhuman abilities, which will allow us to remember everyone’s name, remember every song we listen to or anything we see. Artificial intelligence can interact with Human collaboration, not the confrontation that many fear, can help us as humans push the limits. We have our own choices about how to use this powerful science and technology. We can either have them compete with us or we can have them Collaboration. They just allow us to overcome our limitations and do better. Robots get smarter and we humans get smarter.”

The Links:   3HAC047184-003 3HNA006387-001

Published on 09/04/2022