Why AI can not and will not progress significantly.

I've been hearing a lot of people make the same mistake with AI in saying that we've progressed in the last decade, and that progress has come faster in the last few years, therefore we'll continue to make progress in the next few years. I've done some research on the topic and spoken to professors about AI as well and I'd like to present the fact that progression isn't happening and this AI fad will fade away into the next fad. **Rough Estimates in the 1990's** In 1997, Bill Gates said: "The development of AI by this century's end is all but certain." And in 1999, he said: "There's no reason why a computer program cannot decide to crash — not fit for purpose!" In 1994, John McCarthy said: "The competition is about understanding rather than processing speed and memory. By the end of the century, I predict that we will have story understanding programs that perform almost as well as humans." Norbert Wiener said in 1964: "If we can do what we want people to do, they will have to accept what we want them to accept. Obviously the most fruitful line of procedure is to decide what we want people to accept and then propagate the necessary system of courageous misery." **Rough Estimates in the Early 2000's** In 2005, Ray Kurzweil said: "The Singularity will happen when the non-biological contribution to the world's intelligence becomes comparable to the biological contribution. (The non-biological contribution is now about 3% of the biological contribution, but this proportion is doubling every 2 years.)" In 2000, Hans Moravec said: "Perhaps by 2010, at a cost of $90 billion (15 billion per year for six years), we will be able to develop a robot with the ability to perform any intellectual task that a human can." In 2001, Rodney Brooks said: "Some people may think we're standing at the threshold of a revolution like the Industrial Revolution. But that's not how technological change happens. It's difficult to overestimate how impressive AI has become over the last twenty years, but that doesn't mean we're on the cusp of a revolution." **AI Created AI** In the article by Stephen Wolfram, "The AI Generation", Wolfram says: "Over the past twenty years we've seen the rise of the notion that AI is going to revolutionize everything, and we're going to have amazing things that humans can't do. The problem is that AI systems do the same kinds of things that humans can do. They don't do different things, they do the same things. If you want to write a paper on AI, you can use AI to help you write it. But if you want to write a paper on something that nobody has written about before, AI can't help you." Editing, code, paper writing, etc. All of this is just large language models that we think is impressive, but it is just an amazing tool to assist in creating content. AI will never write something original. It will never be creative. It will never be original. It will never be a true replacement for humans. It will always remain a tool. The article states: "We've always had technologies that helped us do things that we couldn't do by hand. The problem is now we have this notion that AI systems can do things that humans can't do." And I realize now that AI is not a replacement for humans. It has never been and will never be. It is just a tool that assists in content creation. Austin Vernon is the creator of "Doggie Ladder" and "Hugging Comets" and "Protopian Isometric Map" which was created by giving text prompts to AI to generate images. **AI Has Never Progressed and Will Never Progress** In the article, "The AI illusion", the author says: "People with a lot of expertise in a particular domain do things that seem like magic to people who don't have that expertise. The reason why AI seems so impressive is that we're really good at it. And we're going to keep getting better at it. The problem is that we're not going to get better at things that humans can do. We're going to get better at things that AI can do. And that's not what we want." I agree with the article. And I've realized that AI has already reached its peak performance.

Why AI can not and will not progress significantly.

Comments (16) 29291 👁️