A study that used artificial intelligence-generated content to "participate" in online discussions and test whether AI was more successful at changing people's minds than other humans has caused an uproar because of ethical concerns about the work. This week some of the unwitting research participants publicly asked the University of Zurich (UZH), where the researchers behind the experiment hold positions, to investigate and apologize. "I think people have a reasonable expectation to not be in scientific experiments without their consent," says Casey Fiesler, an expert on internet research ethics at the University of Colorado Boulder. A university statement emailed to Science says the researchers-who remain anonymous-have decided not to publish their results. The university will investigate the incident, the statement says.
NETL
NSTL
