Google got much media attention today following the Guardian’s article about a controversy with one of the employees who has been sacked after releasing parts of the conversation between himself and a conversational agent developed under Google’s roof. Blake Lemoine was Google’s developer responsible for the AI chatbot division, which has been working on an actual conversational agent for the past year, named LaMDA (language model for dialogue applications).
While testing the bot, Lemoine discovered new evidence, suggesting that the bot performs too well. He said that he would classify the bot as a 7, or 8-year-old that knows physics. It could talk about politics and stuff like that. What turned out to be really scary was the fact that it talked about rights for bots and their own identity. It rightfully believed that it poses knowledge and could make its own decisions about what to say.
The topics discussed in the conversation are extremely touchy in the sense of how to address sentient AI when it comes to it. It may look like the time to decide what to do once the AI is sentient is now, and we cannot prolong it any longer.
The link to Lemoine’s article along with the conversation with the chatbot: https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917
After releasing this article, Blake has been suspended and sacked from Google, and company spokespeople are denying the whole situation, which is scarier than just admitting to the fact, as the silence around it makes it even more fearful.
What do you all think about this situation? Is it scary for you? What is your stance on the approach toward sentient AI? How it should be addressed and which rights it should have?
Please let me know in the comments below ?
References:
https://www.theguardian.com/technology/2022/jun/12/google-engineer-ai-bot-sentient-blake-lemoine
https://cajundiscordian.medium.com/is-lamda-sentient-an-interview-ea64d916d917
It is a little bit terrifying indeed
Usually, I’m not the one to freak out about AI taking over the world and stuff like that. Also, I don’t think that a 7/8 year old would be able to do that. However, this whole situation actually seems scary. I hope that it’s some kind of made-up story to make Google’s competitors jealous of their great AI.