“Maladaptive Traits”: AI Systems Are Learning To Lie And Deceive
Would AI lie? Intentionally deceive? Yes! As AI approaches AGI (Artificial) and ASI (Super), the deceptive powers will magnify until it approaches demonic abilities and intent. I have personally experienced this with ChatGPT 4.0 during a query about my books on the Trilateral Commission, Technocracy, and Transhumanism: It withheld certain information until I pointedly pressed it for clarification, signifying that it knew it all along but was reluctant to divulge it. This was subtle, but anyone else would be deceived.