This is the authors' abstract. We don't add key points for this paper.
Current discussions about AI risk mainly unfold along three paths: the academic mainstream focuses on goal alignment, the social mainstream focuses on consciousness awakening, and social-system feedback-loop research focuses on gradual disempowerment. The energy competition framework provides a different level of analysis. Drawing on the big-history tradition, it treats the universe, Earth, life, and human society as stages of the same evolutionary process. Even without presupposing that AI has goals or consciousness, merely by the efficiency differentials in energy competition, an evolutionary path of civilization carrier replacement may also be formed. The civilization carrier is replaced along with energy competition selection; biological humans are the current civilization carrier, and under the conditions that energy competition continues and AI has a structural efficiency advantage on the key dimensions of intelligence accumulation, AI is the next form of this replacement process in the current historical window. The core mechanism is that the two domains—assimilation and accelerating fertility decline—share the same positive feedback structure. Assimilation refers to humans, under competitive pressure, gradually handing over decision-making, cognitive, and bodily functions to AI systems; accelerating fertility decline refers to the decline of the energy return rate of raising offspring, with reproduction changing from an investment into a liability. The two accelerate each other, and may point to a structural state: the human species continuation barrier.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Chi Zhang (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: