AI a dream or a reality
A lot of time and research has passed since I wrote my last blog post "The Rise of Artificial Intelligence". I also talked to a lot of people about the topic, which also influenced me. So, let me reconsider my opinion.
Artificial Intelligence Hype
Right now, all AI can do, is to solve some specific tasks very well, as I mentioned in my last blog post. But can it really do anything? I still think, AI is just a buzzword, or more precisely, as mentioned in the article "The Seven Deadly Sins of AI Predictions", by Rodney Brooks, a "Suitcase word". Andrew Ng says "Despite all the hype and excitement about AI, it's still extremely limited today relative to what human intelligence is", what I think too.
Oren Etzioni says, an AlphaGo playing AI might be unstoppable in playing his game, but he could not answer questions like "Is it raining outside" or "Can you tell me about the game?", that's also what Rodney Brooks pointed out in his article.
The article "The Seven Deadly Sins of AI Predictions" shows a big contrast to the article "The AI Revolution: The Road to Superintelligence". I think they're both right in one way and another, but I really don't think that AI will conquer the world in the next few years and kill us all by itself. Don't get me wrong, I'd like to see the Singularity in 2045, as Kurzweil predicts, but according to human mispredictions in history, this won't happen so early. That is also what Rodney Brooks showed when he wrote "Many people are suffering from a severe case of 'exponentialism'."
Artificial Intelligence is a tool
I still have the same view as I wrote in my last blog article in the section about our life with AI, but I would reconsider the year in which Singularity will take place. Technology will come and still will change our way to live and we really rely on technology. Carlos E. Perez think that "Natural Stupidity is more Dangerous than Artificial Intelligence". So, if something stops working, like GPS as mentioned by Rodney Brooks, "we would be cold, hungry, and quite possibly dead".
After learning a lot in Thilo Stadelmann's AI course and knowing how to program a simple AI, I really don't think the AI will ever be self-confident. A lot of data preparation is required for a specific AI and it is very difficult to generalise something like this. For this reason, it will probably take a while until we have the human-like AI. Oren Etzioni says in his TED Talk "AI is neither good nor evil. It's a tool. It's a technology for us to use." We don't need an AI to calculate the physics, for example to predict the velocity of a falling apple. It is a tool to help us in the area that we don't know exactly. But we should still use AI responsibly.
"What all of us have to do is to make sure we are using AI in a way that is for the benefit of humanity, not to the detriment of humanity." - Tim Cook