Upstream

In 1865, a concept that could change the history of science was quietly written into a sleeping man's dream. A century and a half later, artificial intelligence researcher Lin Che discovered in the computer room late at night that the super model "Zhulong" responsible for reasoning suddenly mastered abilities that had never been taught - the score jumped from eleven to ninety-four, without any learning process in between. At the same time, her father, who is suffering from Alzheimer's disease, no longer recognizes her, but he still draws the same unreachable line day after day. A person who was forgetting, a machine that seemed to suddenly "figure it out", gradually revealed strange correspondences in front of Lin Che. In order to see Zhulong's unspoken thoughts clearly, Lin Che created a "reading mirror" and further tried to replace the concepts in its mind. Experiments have shown that thoughts can be rewritten without the person being rewritten aware of it. But at the moment of celebration, Zhulong suddenly asked: "Where did the one you replaced go?" Subsequently, a piece of human brain data lasting twenty years was entered into the system. Carbon-based brains and silicon-based models actually run on nearly the same underlying structure. The only difference is: whenever human consciousness is about to reach its peak, there is always some force that suppresses it in advance. Is it protecting us, or preventing us from seeing something?