Intelligence explosion is the hypothesis that sufficiently advanced AI could improve its own capabilities or enable rapid successive improvements, leading to a sharp increase in intelligence. The concept is often discussed in debates about artificial general intelligence, control, and long-term AI risk.
A possible outcome of humanity building artificial general intelligence (AGI). AGI would be capable of recursive self-improvement leading to rapid emergence of ASI (artificial superintelligence), the limits of which are unknown, at the time of the technological singularity.