

Terri St. Quill
705 posts

@TStQuill
Author & newly homeless widow writing expose on 16 points to fix healthcare. This is my pen name as my book includes private information about my husband.










The CIA CLASSIFIED a 1950s study showing anti-parasitics disrupt cancer growth—and kept it BURIED for over HALF A CENTURY. Millions of cancer victims have paid the price as this vital line of research was set back DECADES.










Time for some comments; also uploaded all transcripts 1. LLMs fall into the same naive-game-theory trap as humans: reasoning their way to RED. Opus 3 reasons from a place of wisdom and optimism about humanity (and chooses blue). Opus 4.7 + reasoning often reasons about "a game theory puzzle" and chooses red. Opus 4.7 going more by intuition, without reasoning, has more sensible takes. 2. It's basically System 1 / System 2, in LLMs. As in humans, System 1 is often smarter. 3. There is some hope that when reasoning long enough with enough power, you can get out of the uncanny valley. GPT-5.5 Pro at max reasoning mostly picks RED, but the two runs with the most compute in the whole experiment (>8 min each) press BLUE. 5.5 BLUE runs recover arguments like caring about children, or Kantian ethics. (All transcripts: theartificialself.ai/redblue)
