When deploying artificial skills, decision-makers often assume that layering human oversight is a safe harbor that mitigates the risks of full automation in high-complexity tasks. This paper formally challenges the economic validity of this widespread assumption, arguing that the true bottom-line economic utility of a human-machine skill policy is highly contingent on situational and design factors. To investigate this gap, we develop an in-silico exploratory framework for policy analysis based on Monte Carlo simulations to quantify the economic impact of skill policies in the execution of tasks presenting varying levels of complexity across diverse setups. Our results show that in complex scenarios, a human-machine strategy can yield the highest economic utility, but only if genuine augmentation is achieved. In contrast, when failing to realize this synergy, the human-machine approach can perform worse than either the machine-exclusive or the human-exclusive policy, actively destroying value under the pressure of costs that are not sufficiently compensated by performance gains. This finding points to a key implication for decision-makers: when the context is complex and critical, simply allocating human and machine skills to a task may be insufficient, and far from being a silver-bullet solution or a low-risk compromise. Rather, it is a critical opportunity to boost competitiveness that demands a strong organizational commitment to enabling augmentation. Also, our findings show that improving the cost-effectiveness of machine skills over time, while useful, does not replace the fundamental need to focus on achieving augmentation when surprise is the norm, even when machines become more effective than humans in handling uncertainty.