Dean Ball, Head of Strategic Futures at OpenAI, suggests that frontier AI laboratories may evolve into "new kinds of institutions" capable of acting as a counterbalance to government power. As AI transitions from specialized tools to foundational infrastructure for cognitive labor—impacting sectors such as cybersecurity, scientific research, and military analysis—the traditional regulatory hierarchy between the state and private corporations may dissolve. This shift occurs because governments are becoming increasingly dependent on privately controlled AI for intelligence, public administration, and critical infrastructure. However, unlike democratic states, AI companies lack electoral accountability, transparency via freedom-of-information laws, or judicial oversight. If AI becomes indispensable, it may necessitate new regulatory frameworks similar to those governing utilities or banks, including continuity requirements and independent auditing. Furthermore, the shift of economic activity from human labor to machine intelligence poses fiscal challenges, potentially requiring novel taxation models, such as a levy on commercial AI inference tokens, to replace lost income and payroll tax revenues. Ultimately, the rise of AI labs as institutional actors necessitates a reevaluation of how to balance the power of non-democratic entities that possess the capability to influence global economic and social processes.