Last year at Google I/O, the Language Model for Dialogue Applications or its abbreviation LaMDA was shown. It is Google's latest neural network that will allow future virtual assistants to initiate and continue conversations like real humans. LaMDA is trained to better understand its conversations with humans which is then used in Google Assistant to operate smarter.
But according to a report in The Washington Post, Blake Lemoine, an engineer at Google, has been ordered to take time off after accusing LaMDA is now an artificial intelligence that has reached the intelligent stage (sentient). According to him in his conversation with LaMDA, he claimed to have feelings and emotions. He added that the capabilities shown by LaMDa had never been seen in the chatbot software that had been developed before.
Google denies his accusations because technology and corporate ethics experts have found no evidence that LaMDA has reached the stage of cognition as indicted. But this neural circuit now has the ability to carry out conversations much like an intelligent system.
This Google reality was approved by artificial intelligence experts because the neural circuits taught to carry out realistic conversations were not powerful enough to reach the stage of reasoning as claimed by Lemoine. Because Lemoine published a conversation with LaMDA publicly without Google approval, he has now been directed to take a leave of absence.
Artificial intelligence that reaches the stage of being intelligent is often a delusion in science fiction works. When Skynet reaches this stage, it launches a nuclear attack on humans in the Terminator series. In The Matrix, high artificial intelligence builds their own country to be free from human shackles before fighting and enslaving humans.