Evolutionary Computation and AI Safety: Research Problems Impeding Routine and Safe Real-world Application of Evolution
This addresses safety concerns for researchers and practitioners deploying evolutionary AI systems, but it is incremental as it builds on existing AI safety frameworks.
The paper tackles the problem of integrating AI safety principles into evolutionary computation to prevent harmful accidents in real-world applications, by exploring how safety issues arise in evolutionary systems and how insights from evolutionary computation can inform broader AI safety research.
Recent developments in artificial intelligence and machine learning have spurred interest in the growing field of AI safety, which studies how to prevent human-harming accidents when deploying AI systems. This paper thus explores the intersection of AI safety with evolutionary computation, to show how safety issues arise in evolutionary computation and how understanding from evolutionary computational and biological evolution can inform the broader study of AI safety.