About 40-50% of Americans believe in some form of ESP. (I am not among them). About 30% believe in telepathy.
I remember coming across the Journal of Parapsychology at my college library in the early 1980s. The research never questioned whether ESP existed, but instead investigated the conditions under which it occurred. Belief in ESP was more widespread in the past. Turing's 1950 paper addressed many of the objections to his proposal that machines could one day pass for human intelligence. One of the objections was that if the judge was able to communicate with the human player by telepathy, then the computer would fail the test. He proposed separating the players in a telepathy proof room if necessary. The gambling industry has disproved the existence of ESP in trillions of experimental trials on billions of human subjects. That's enough evidence for me. On Tue, Sep 29, 2026 at 2:18 PM John Rose via AGI <[email protected]> wrote: > On Tuesday, September 29, 2026, at 12:12 PM, Matt Mahoney wrote: > > People like to sound smart by throwing together lots of big words to express > grammatically correct nonsense. > > Isn't that a form of compression? Depending on the receiver. Yes. An LLM will do that. If it is trained on input like "quantum consciousness" it will learn to associate the words, increasing prediction accuracy. It doesn't matter if the relationship is false. An LLM's model of the world is made up of what people believe about the world. If you only train AI on human knowledge, then it will only be as smart as humans. And we won't be smart enough to tell if we were successful at making it any smarter. The top AIs know as much as 20,000 people. But that is a long way from 8 billion. -- -- Matt Mahoney, [email protected] ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T906c878e4ba411be-M060cd1a359a6f33c7c7cd26e Delivery options: https://agi.topicbox.com/groups/agi/subscription
