This conceptual study examines the use of agentic artificial intelligence (AI) for autonomous decision-making in stock market investments. Agentic AI systems possess independent action capabilities and strategic reasoning that can enhance investment performance in dynamic financial markets. The study proposes a theoretical framework that integrates AI autonomy, decision-making theory, and investment strategy to explain how agentic AI can support business investment decisions. A central emphasis is placed on model retraining to counter data drift and maintain predictive accuracy, keeping AI systems aligned with changing market conditions. The framework identifies key design, governance, and deployment considerations for autonomous AI in financial environments. No empirical validation is undertaken in this study, and the findings highlight that continuous, intelligent model retraining will be crucial for future work to achieve sustainable, reliable AI-driven investment decision-making.