Interesting experimental logical AI has promising implications for AI Safety.
Claiming to have created a language that enables developers to create software and AI that can reason over its own future versions.
Original post: https://x.com/TauLogicAI/status/1841813606154793354
Abstract Summarizing their process and the language tech: https://tau.net/Logical-AI-Software-Specification-Reasoning-GSSOTC.pdf
Full paper: https://tau.net/Theories-and-Applications-of-Boolean-Algebras-0.25.pdf
But the full paper is super long and goes over my head, the abstract is much easier to digest.
New AI Software Specification Language Could Solve AI Alignment and AI Safety For Future AI and Machines
byu/TheBlueRefinery29 inFuturology