In the Name of “Efficiency”: The Practice and Controversy over the Embedding of AI in the DOGE Reform

With the rapid evolution of artificial intelligence (AI), smart technology is increasingly becoming a key variable in reshaping the organizational forms and power structures of modern government. This study takes the U. S. Department of Government Efficiency (DOGE) as a case to systematically investigate the concrete paths through which AI has become embedded in governance and the controversies and conflicts it has triggered. The research posits that the DOGE reform is not merely a technical adjustment aimed at enhancing administrative efficiency but also a governance experiment with technology at its core, profoundly reflecting the intrinsic tensions between technological rationality and institutional logic. The study finds that DOGE promotes the structural embedding of AI into the federal administrative system primarily through a threefold path: first, through data control to achieve the centralized dispatch of cross-departmental information and the integration of resources; second, through process re-engineering to reshape decision-making by deploying algorithmic models in areas such as personnel adjudication and regulatory review; and third, through institutional adaptation, relying on executive orders to neutralize the legal and procedural constraints caused by the deployment of AI, thereby providing a sanctuary of legitimacy. However, this aggressive embedding process has triggered multi-dimensional conflicts and diverse forms of resistance. The core controversies are concentrated in three areas: security and privacy risks arising from the monopolization of data, accountability deficits caused by the algorithmic black box, and the erosion of public values due to the logic of performance quantification. In response, various actors have initiated countermeasures through different pathways: the judicial system and Congress have challenged its legitimacy and compliance through lawsuits and oversight; the bureaucratic system has weakened its policy effectiveness through procedural delays and selective compliance; professional groups have resisted its technological encroachment on public services through mass resignations and public statements; and civil society has criticized its impact on privacy, labor rights, and democratic mechanisms through public opinion, media, and social organizations. The DOGE case clearly reveals both the immense potential as well as the latent perils from the embedding of AI in governance. This reform experiment has given rise to a technocentric administrative paradigm that bypasses the traditional bureaucracy, challenging the existing power boundaries and accountability logic within the power structure and eroding the pluralistic values embedded in public service. Therefore, it is necessary to approach the challenges and risks of AI-embedded governance with caution and to place greater emphasis on maintaining a dynamic equilibrium among institutional design, public values, and responsible ethics.