Any normal person who has tried using so-called ‘AI’ for a period of time can tell you that there’s something “off” about it. For example, ChatGPT is supposed to be the gold standard among AI large language models (LLMs). If ChatGPT is really an artificial intelligence, why is it incapable of knowing who the current president is? Ask it right now and it will tell you Joe Biden is still president. Try it.
Since we’re restructuring the entire American economy around this new piece of software, shouldn’t accuracy and truthfulness be a goal to try to attain? These so-called ‘AI’ programs don’t seem capable of that.
Rather than simply saying, “I don’t know” when it encounters a question it can’t answer, all of these LLMs make up plausible-sounding BS out of thin air. They’ll invent articles, publications, and even court cases, and then invent quotations from those fictitious documents just to satisfy the person on the other end of the search query.
Karl Denninger is a businessman and programmer. For those who don’t remember him, Karl founded MCSNet in the 1980s and was a founding member of the Tea Party movement in America in 2008.
Denninger has a lot of interesting insights into AI that you won’t see reported on any mainstream news outlet. He’s considered “dangerous” by the establishment because of his opinions. In fact, Google forces him to keep most of his videos on his own website behind a paywall.
Here’s one that is definitely worth watching, with his thoughts on so-called ‘AI’: