Artificial Intelligence in Medicine and Surgery - An Exploration of Current Trends, Potential Opportunities, and Evolving Threats, Volume 3

Artificial intelligence (AI) and machine learning (ML) have become among the most powerful forces shaping our civilization. Despite various predictions of “doom and gloom,” the rumors of humans “becoming replaced” by artificially intelligent machines are slow to materialize. This discrepancy between predictions and the observed reality most likely reflects the persistent importance of the “human f...

Artificial intelligence (AI) and machine learning (ML) have become among the most powerful forces shaping our civilization. Despite various predictions of “doom and gloom,” the rumors of humans “becoming replaced” by artificially intelligent machines are slow to materialize. This discrepancy between predictions and the observed reality most likely reflects the persistent importance of the “human factor,” or the simple fact that there are no substitutes for human empathy, intuition, and creativity, especially in the context of complex real-life scenarios and cognitive challenges. In addition to strong support for the viewpoint that AI is unlikely to replace humans, but rather “humans using AI may replace those who do not,” it is quickly becoming evident that effective AI design, implementation, and operations may be less successful, if not severely impaired, without ongoing human oversight and human content expert input. Thus, the overall AI/ML-based paradigm continues to evolve toward human-AI synergy and the gradual evolution of human-centric applications and processes. We term this paradigm “modern world plus AI,” and it is based on the reality that artificial intelligence is a human invention and inherits the very same shortcomings and biases as its creators. It is also reasonable to propose that for AI to reach its fullest potential, it must be inherently synergistic with humanity. Consequently, an honest and open discussion at the global level should take place, ensuring that AI-based systems are implemented in accordance with universally accepted social and behavioral standards and expectations, including fundamental respect for humanity and human values, and that corresponding ethical frameworks are grounded in a human-centric agenda. Robust and “fool-proof” mechanisms are needed to prevent systematic biases from trickling into AI models across all applicable operational domains. As we continue to examine any potential benefits, risks, and alternatives related to AI/ML implementations – both in healthcare and beyond – we must make every effort to ensure that we create a future where humans actively decide and shape this “modern world plus AI” evolutionary and synergistic paradigm, and that human content experts and overseers have primary and overarching input into the entire spectrum of AI/ML-based system design, implementation, and practice-based improvement. Finally, we must remember that the “AI revolution” creates a unique opportunity to achieve rapid global parity, where “no one is left behind” and where humanity is given a chance to participate and develop to the fullest.