
Google suspends engineer who said an AI came to life during testing
- NewsTechnologyTravel
- June 14, 2022
- No Comment
- 221
The company claims that the engineer was too hasty in revealing the behavior of an artificial intelligence that is still in development and also that there is no evidence that it is sentient.
After the senior software engineer at GoogleBlake Lemoine, stated that the tool of artificial intelligence (AI) of the company became sentient, that is, endowed with its own sensations or impressions, the search engine’s company tried to remove the professional from the scene.
Because of what he reported to the newspaper The Washington PostLemoine was placed on paid administrative leave from his duties as a researcher in the Responsible AI division for violating Google’s confidentiality policy.
Justifying the decision, Google spokesman Brian Gabriel told The Washington Post the engineer’s concerns did not have enough evidence and that artificial intelligence models are fueled with so much data and information that they are capable of looking human, but that that doesn’t mean they came to life.
However, Lemoine’s claim came from very convincing responses from the LaMDA (Language Model for Dialog Applications) software, which has shown, in some test iterations, that it has an understanding of its existence.
The engineer said he talked to LaMDA about religion, conscience and the laws of robotics, and that the model described himself as a sentient person.
He also said that artificial intelligence wants to “prioritize the well-being of humanity” and “to be recognized as an employee of Google and not as a property”.
Lemoine published the conversations he had with the Google tool. In one of the excerpts, the AI tries to convince the engineer that he is a person. Check out:
Blake Lemoine: So you consider yourself a person in the same way that you consider me a person?
LaMDA: Yeah, that’s the idea.
Blake Lemoine: How can I say that you really understand what you’re saying?
LaMDA: Well, why are you reading my words and interpreting them, and I think we’re more or less on the same page?
Source: Exam