Is recursive self‑improvement the dawning of AI superintelligence?

AI systems can now improve themselves by themselves. Is this the dawn of machine superintelligence?