Washington –
Experts have long warned about the threat posed by artificial intelligence, but a new research paper suggests it is already happening.
Today’s AI systems, designed to be fair, have developed a disturbing knack for deception, from tricking human players in online games of world conquest to hiring humans to run prove-you-not-a-robot tests. solve, a team of scientists claim in Patterns magazine on Friday.
And while such examples may seem trivial, the underlying problems they reveal could soon have serious real-world consequences, says first author Peter Park, a postdoctoral researcher at the Massachusetts Institute of Technology who specializes in AI-based existential security.
