The creation of artificial moral agents (AMAs) capable of moral intuition and reasoning offers a promising strategy for reducing algorithmic moral biases and improving AI alignment. Yet, the very idea of AMAs has been criticized, either for being technically implausible or for being normatively undesirable. This chapter addresses both forms of skepticism by outlining and defending a model of instrumental moral reasoning that is compatible with the capacities of current advanced AI systems. The model aims to replicate within artificial agents an improved (i.e., objectively aligned) form of moral intuition and reasoning. I will call this approach to AMA enhanced intuitionism. My argument proceeds as follows. After introducing the rationale for AMAs (Sect. 10.1), Sect. 10.2 maps the main competing approaches to AMAs, distinguishing between behavioral and cognitive models, and between commonsense-oriented and ideal-morality-oriented perspectives. Section 10.3 develops the enhanced intuitionist framework, which is both cognitivist and grounded in commonsense morality. Building on this conceptual foundation, Sect. 10.4 outlines a cognitive architecture for reproducing moral intuitions and instrumental moral reasoning within artificial systems. Section 10.5 addresses several normative objections to AMAs and evaluates whether the creation of such agents is, all things considered, desirable.