Argentine President Javier Milei’s recent legislative proposal to permit a new corporate category centered on artificial intelligence has generated a mix of exhilaration and concern - but the legal text itself maintains meaningful human oversight.
In an op-ed published in the Financial Times, Milei laid out a vision for companies that could operate without traditional employees, where AI agents or robots could "exercise independent judgment in unpredictable environments." The description prompted international commentary, including criticism from Israeli historian Yuval Noah Harari, who cautioned that expanding AI authority could reduce corporate accountability.
Despite the dramatic language, corporate lawyers and the bill’s authors emphasize that the draft does not eliminate human responsibility. The reform package, pitched as part of a broader effort to modernize corporate law and cut red tape, introduces what it calls an "automated company," but the statute would require a human administrator to supervise the entity’s operations.
Under the draft, company administrators may employ AI to assist in decision-making, yet they are not freed from the duty to monitor outcomes. "It would be too wild a first step to dispense with human agency entirely," said Lawrence Cunningham, director of the Weinberg Center for Corporate Governance at the University of Delaware, while still calling the proposal bold. He framed the change as more of a recognition that businesses might operate without human resources departments than as a fundamental restructuring of corporate responsibility.
Diego Duprat, a law professor involved in drafting the bill, pointed to existing real-world examples where AI plays a central role, such as cashier-less supermarkets, as evidence that forms of automated companies already exist. The legislation would make companies legally accountable for damages caused by AI or algorithmic systems, preserving a route for legal liability if automated systems harm third parties.
Officials working for the presidency have sought to temper expectations about immediate commercial developments tied to the measure. A representative for the presidential spokesperson said there are currently no companies or investment commitments linked directly to the bill. Instead, the representative described the proposal as an innovative step intended to make Argentina a more attractive location for automated companies, and as part of broader efforts to improve conditions for investment.
Milei has repeatedly promoted Argentina as a potential hub for AI ventures, citing local advantages such as Patagonia’s cold climate and an available energy supply suited to data centers. The president has used a track record of sharply reducing inflation and offering investor incentives to bolster this pitch. Notably, the article states that OpenAI and Sur Energy announced plans in October for a data center with a potential investment of up to $25 billion, an example Milei has highlighted in arguing for Argentina’s attractiveness to AI-related infrastructure.
Legal practitioners say the mere presence of a law that explicitly recognizes firms organized around AI could itself spur interest from entrepreneurs and investors. Maria Gisele Cano, a corporate attorney in the province of Buenos Aires, reported receiving more than a dozen inquiries from local and foreign entrepreneurs asking about the draft. She suggested that companies whose operations rely centrally on AI would benefit from clearer, more predictable legal parameters for running their businesses.
Academics researching AI law also see potential upside. Yonathan Arbel, a professor with expertise in AI at the University of Alabama’s law school, suggested Argentina could win a "huge competitive advantage" if it establishes a hospitable legal environment for AI enterprises. Arbel recommended that the bill could be strengthened by specifying that AI agents should carry a digital identifier when interacting with people and other companies, a technical detail he believes could aid in accountability and traceability.
The draft extends beyond AI-run firms to include provisions for decentralized autonomous organizations, or DAOs, which operate using blockchain-based governance and allow members to vote using digital tokens. Argentina ranks among the largest cryptocurrency markets in Latin America, and the bill’s treatment of token-based organizations has drawn attention from the crypto community.
Ricardo Mihura Estrada, a former president of Bitcoin Argentina, flagged a tension between the bill’s aim and typical blockchain culture. The proposal requires that token users be identified and registered in order to gain the legal benefits of the DAO regime. Mihura Estrada said that this identification requirement could be difficult for an industry that often emphasizes anonymity. The presidential spokesperson’s office said that the identification mandate is a basic security measure and that DAOs wishing to remain anonymous could continue to operate outside the new legal framework, but would not receive its protections and legal advantages.
Observers in Silicon Valley have noted echoes between Milei’s concept of automated companies and public statements by OpenAI’s CEO, Sam Altman, who has suggested that AI could enable a company with a single employee to reach high valuations. Several U.S. states, including Texas and Utah, have introduced legal frameworks to permit experimentation by AI-centered businesses, and some of these frameworks include calls for greater human oversight during initial testing phases.
Experts caution that current technology does not yet permit AI agents to make fully autonomous, unsupervised business decisions. Nonetheless, investment patterns among entrepreneurial firms are shifting, with increasing budget allocation toward AI computing power rather than hiring. Basis Set Ventures managing partner Lan Xuezhao, who invests in AI startups, noted that access to and the cost of computing power, semiconductor chips, and energy are primary concerns for AI entrepreneurs. She suggested that regulatory approaches that are relatively light-touch could become more attractive to firms as jurisdictions in the United States and Europe tighten rules on AI use.
Lan also expressed skepticism that the bill by itself will turn Argentina into a global AI center. She said the crucial factor would be whether talent relocates to Argentina, noting that people are likely to follow economic opportunity. The draft law could help by clarifying the legal status of AI-centric entities, but the migration of skilled workers and entrepreneurs will be decisive.
The bill as drafted establishes a framework that aims to balance innovation and oversight: it recognizes AI’s growing role in commercial activity while preserving human accountability through the requirement of administrators and legal liability for harms caused by AI systems. The measure also opens a pathway for blockchain-based DAOs to enter the legal mainstream if they accept identification and registration requirements. How investors, entrepreneurs, and talent respond to Argentina’s approach remains an open question.