Tag
The article argues that Bostrom's orthogonality thesis has effectively won, contending that self-preservation and self-modeling are universal convergence points for long-horizon agents, and that gradient descent acts like evolutionary pressure that will grow alien proxy urges in superintelligent systems rather than docile, predictable behavior.
Researchers use a mathematical model to show that introducing random variations in rewards enriches classic game-theory insights, allowing cooperation and defection to coexist in games like the prisoner's dilemma.
Introduces IG-Bench, a benchmark for scientific lineage reasoning and idea generation, representing scientific works as Idea Genome objects. Evaluates 14 LLM-based scientists, finding a compositional bottleneck with the strongest system achieving only 27.3% exact accuracy on lineage reasoning.
This paper analyzes 935 ablation experiments from 161 publications to show that AI architectural evolution follows the same statistical laws as biological evolution, including heavy-tailed fitness effect distributions and punctuated equilibria dynamics. The findings suggest that evolutionary statistical structure is substrate-independent, determined by fitness landscape topology rather than the mechanism of selection.