arXiv:2608. 05558v1 Announce Type: cross Abstract: This paper examines Turing's 1948 report, "Intelligent Machinery", as an important conceptual source for the later imitation games.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.
The paper analyzes Turing’s 1948 report "Intelligent Machinery" as a foundational source for later imitation games, highlighting key design concepts such as the possibility of machine errors, the exclusion of irrelevant physical traits, the role of a human judge, and Turing’s view that intellectual activity is largely search. It argues that limiting the human contestant to a weak chess player heightens the importance of intellectual search, making human behavior more comparable to machine behavior. This reframes the 1948 game as a human‑approximates‑machine scenario, suggesting that imitation games can probe when human intelligence becomes machine‑like under specific task constraints.
arXiv:2501. 17629v2 Announce Type: replace-cross Abstract: Several studies claim that large language models have passed the Turing Test and hence can "think", yet none follow Turing's original instructions precisely.
The article proposes an updated Turing Test that incorporates energy consumption as a key metric, arguing that the original test is insufficient for distinguishing human from machine intelligence in the context of modern AI. It suggests that by adding an energy constraint, the test evaluates intelligence through the lens of efficiency, linking abstract thinking to tangible resource limits. The new test also provides a measurable, practical endpoint, encouraging society to balance AI time savings against total resource costs.
arXiv:2605.10851v2 Announce Type: replace Abstract: We initiate the study of the Generalized Turing Test (GTT), a formal generalization of Turing's imitation game from humans to arbitrary interactive...
The paper investigates whether large language model (LLM) agents can design AI methods that outperform or differ from human-designed approaches. By mapping both human- and agent-designed methods into task‑specific algorithmic design spaces, the authors evaluate performance and algorithmic differences across multiple modalities. Results show that while agents occasionally match or exceed human state‑of‑the‑art performance, 96.8% of their designs fall within human‑derived spaces, often recombining or exactly matching existing human algorithms.
arXiv:2510. 05743v3 Announce Type: replace Abstract: We review the historical development and current trends of artificially intelligent agents (agentic AI) in the social and behavioral sciences: from the first programmable computers, and social simulations soon thereafter, to today's experiments with large language models.