The Pentagon is confronted with an ethical dilemma concerning the use of artificial intelligence (AI) in combat: should there be full human supervision, which may lack speed and accuracy, or should AI operate without any human control, which could lead to unethical behavior. These authors propose a novel approach, the “AI operator,” which facilitates a synergistic relationship between military personnel and AI without compromising ethics. This solution draws inspiration from the historical use of war elephants in combat, where a dedicated operator managed and directed these creatures. AI operators, like these historical trainers, would oversee the AI’s development, training, and deployment, while also maintaining ethical considerations. 

Key Points: 

1. The use of AI in warfare presents a challenge between maintaining ethical standards via human oversight and the potential for faster, more accurate unsupervised AI operations.
2. The concept of an “AI operator” is suggested, which models after the historical role of a mahout overseeing a war elephant, thus maintaining a balance between human oversight and AI efficiency.
3. The AI operator would bridge the gap between military personnel and AI, understanding the nuances of the AI system, and ensuring that ethical standards are met.
4. To prevent model biases, diverse teams should create AI models using unique data sets and training environments rather than tweaking and customizing existing models.
5. Embracing this collaborative approach between humans and AI would turn the commitment to ethical values from a perceived limitation into a strategic advantage and can be extended beyond the battlefield into other applications where ethical considerations are important.

Continue reading at https://www.c4isrnet.com/opinion/2023/06/14/what-war-elephants-can-teach-us-about-the-future-of-ai-in-combat/