Lots of links on LaMDA

Abstract

The article discusses the case of Blake Lemoine, a Google engineer who claimed that LaMDA, a large language model, was sentient. The author presents a variety of perspectives on Lemoine’s claim, including those from prominent AI researchers and ethicists, as well as a critical examination of the evidence presented by Lemoine. The author argues that Lemoine’s claim conflates different questions about AI, such as intelligence, consciousness, sentience, and personhood. Additionally, the author highlights the risks of conflating these concepts and the potential for AI companies to exploit this confusion. The article concludes that, while the question of AI sentience is important, it is crucial to avoid oversimplification and to focus on the actual risks of AI, such as bias and fairness, as well as the potential for AI systems to be used for harmful purposes. – AI-generated abstract.