Artificial superintelligence
Artificial superintelligence (ASI) is intelligence that vastly exceeds the best human minds across essentially all domains — scientific, creative, and strategic — and could improve itself recursively.
ASI is the capstone and the greatest risk: it could solve problems beyond human reach, but only if built to remain aligned with human values — the core of ASIZ's safety mission.
Entirely prospective; today's systems are far from it, and how to build and control it safely is unsolved.
The discovery loop, applied.
Through the Computational Discovery center, ASIZ aims the discovery loop at artificial superintelligence: frontier models survey the literature and simulate the possibilities, rank the experiments most likely to resolve the open questions, and partner labs run the decisive ones — with results shared openly.