IS THE future safe from a robot dystopia? Stephen Hawking and Microsoft research director Eric Horvitz aren鈥檛 convinced. They have signed an open letter offering cautious advice for the field of artificial intelligence, which was published this week by the Future of Life Institute, a research organisation in Boston.
The letter highlights the potential for AI to do good, but argues that several steps must be taken to ensure that future machines are beneficial and reliable. For instance, researchers need to show that their creations work as intended and are safe from cyberattacks. They should also explore how to maintain 鈥渕eaningful human control鈥� over intelligent machines, in order to avoid 鈥渦ndesired consequences鈥�.
鈥淚 think it鈥檚 very important that everybody knows that AI researchers are seriously thinking about these concerns and ethical issues,鈥� says Francesca Rossi, a computer scientist at Harvard University, and one of the first to sign the letter. 鈥淲e have to make sure that machines will behave in a way that we think is right.鈥�
Advertisement
This article appeared in print under the headline 鈥淏eware AI dangers鈥�
