Scientist says 'Everyone on Earth will die' if AI is allowed to be more intelligent

tech.hindustantimes.com:

We are entering an important era of technology — the rise of AI. Artificial intelligence has been among us for a long time, but recent developments have pushed its capabilities and intelligence to a level where it may start to leave humans behind. A prime example of this is OpenAI-build ChatGPT. Based on the GPT-4 language processing model, this AI chatbot is capable of processing vast amounts of data, analyzing it and thereafter generating content. In fact, it can answer just about any question asked of it. But this is just the beginning. An AI researcher has recently claimed that “literally everyone will die” if AI is allowed to grow more intelligent without any checks.

Eliezer Yudkowsky, of the Machine Intelligence Research Institute in Berkeley, California, told The Sun, “Many researchers steeped in these issues, including myself, expect that the most likely result of a superhumanly intelligent AI is that literally everyone on Earth will die”. And he thinks this is an inevitability. “Not that this might be possible, but that it would be the obvious thing to happen,” he added.

But why exactly does he fear AI so much? Let us take a look.

On the surface, an AI chatbot is a greatly helpful tool that can improve the efficiency and productivity of humankind. It can act as a direct source of information available on the internet so users do not have to spend time scrolling through pages. It can also analyze large amounts of text to let users know about a precise data point. Recently, ChatGPT even helped diagnose an illness in a dog correctly and saved his life, leaving vets stunned.

But all that is the good stuff. There is a dark side to this as well. And it is called misinformation, deep fakes, data privacy issues, security