Tag
This arXiv paper introduces a network-derived 'strategic tension' metric based on spectral radius to compare human and AI decision-making in Go, finding that while tension decays with human skill, elite play after AlphaGo's 2016 breakthrough shows a surge in sustained tension, framing AI as cognitive scaffolding that elevates human handling of complexity.
Beacon shares its research vision for Company World Models (CWM): AI systems that observe business activity, hypothesize and implement strategic improvements by dispatching digital employees, and measure outcomes to drive company-level decision making. The post outlines the capability and alignment challenges involved and argues Beacon's structure enables the fastest research path toward this goal.
A free open-access textbook on game theory, with 585 pages and 165 solved exercises, is promoted as a resource for economics, AI, computer science, and more.
This paper introduces CEO-Bench, a multi-agent benchmark for evaluating LLMs on CEO-level strategic resource reallocation, revealing systematic failure modes and a structural integration–boldness tradeoff.