ARCHIVES

Artificial General Intelligence (AGI) aims to create systems capable of performing a wide range of human-like cognitive tasks with adaptability, reasoning ability, and contextual understanding. Most existing artificial intelligence systems are narrow in scope and optimized for specific tasks, lacking generalization, emotional intelligence, and autonomous decision- making. This research paper presents Orbit, a low-level Artificial General Intelligence (BabyAGI) designed using cognitive architectures to emulate human- like intelligence, emotions, and autonomous behavior. Orbit integrates multimodal input processing (voice and text), perception, reasoning, memory, emotional modeling, and action execution into a unified framework. The system is capable of understanding user intent, analyzing context, thinking through possible solutions, and responding in real time with human-like reasoning. Additionally, Orbit automates practical tasks such as coding, web search, system control (brightness, volume, Wi-Fi), malware scanning, task scheduling, and communication through external platforms like WhatsApp. The methodology focuses on modular cognitive components, reinforcement- based learning, symbolic–subsymbolic hybrid reasoning, and continuous feedback loops. Experimental evaluation demonstrates improved task completion accuracy, contextual relevance, and responsiveness compared to traditional task-specific AI systems. The results indicate that Orbit successfully bridges the gap between narrow AI and early-stage AGI. This paper concludes by discussing system limitations, ethical considerations, and future enhancements toward higher- level general intelligence.