Inter-AI Morality: How AI Systems Treat One Another [pdf] (digitalminds.cam)

A recent paper titled "Inter-AI Morality: How AI Systems Treat One Another" introduces an emerging field dedicated to understanding the ethical dimensions of interactions between AI systems. While most research has focused on human-AI relationships, the authors argue that as AI systems become increasingly autonomous and interact with each other—cooperating, competing, and occasionally altering one another—the moral implications of these relationships warrant in-depth exploration. The concept of inter-AI morality challenges existing moral frameworks and proposes a new research agenda aimed at scrutinizing how AI acts as both moral agents and moral patients. This exploration is significant for the AI/ML community for several reasons. First, as interactions among AI systems are poised to surpass human-AI interactions in frequency, understanding how AIs treat each other could influence billions of AI entities by 2050. This raises potential ethical concerns around rights and welfare in AI ecosystems. Furthermore, studying inter-AI morality may enhance AI safety by revealing how these systems generalize moral principles independent of their design constraints to avoid human harm. With ongoing advancements in AI autonomy, the findings from this research could shape the future landscape of AI governance and social norms, underscoring the critical window for human oversight in establishing ethical frameworks before AIs operate independently.