Possibility of Imposing Artificial Intelligen...

Possibility of Imposing Artificial Intelligence with Full Human Characteristics

Author Name : Ms. Sangita Dey

DOI: https://doi.org/10.56025/IJARESM.2023.1152394

ABSTRACT   Advance AI research and an increase in AI applications in decision-making processes, security-critical domains, and ethically relevant frames, a large-scale debate on potential safety measures encompassing corresponding long-term and short-term issues has emerged across various disciplines. One relevant question in this context that has been addressed by several AI Safety researchers is, for example, the AI alignment challenge, on which no definitive consensus has yet been reached. Despite significant advances in robotics and artificial intelligence, robots are still essentially zombies that perform actions and tasks without comprehending what they are doing. Deep-Learning AI programmes classify vast volumes of data without understanding the meaning of their inputs or outputs. We still need a real understanding of the underlying concepts and procedures that would allow robots to comprehend their surroundings, be aware of what they are doing, take appropriate and timely initiatives, learn from their own experience, and demonstrate that they are aware of what they have learned and how. The research will investigate into whether it is possible to impose human-like consciousness onto AI.