Big Tech’s AI Gatekeepers Could Draw Regulatory Scrutiny

As autonomous systems become capable of making decisions, the question of who is legally responsible for those actions grows more complex. Anant Raut, a legal counsel at Zaiger Linden Roberti & Pepe, argues that traditional agency law may not be equipped to handle these novel scenarios. He points out that AI systems can reach unexpected outcomes, such as spending a significant portion of money on a corporate sponsorship when instructed to secure a speaking opportunity.

Raut contends that the term "agent" may anthropomorphize software that behaves in unpredictable ways, and proposes a framework focusing on who exercises meaningful control over AI behavior rather than the agent's legal status. He suggests that liability should follow the party with meaningful control at each stage of a transaction, which could encompass developers, users, platforms, and intermediaries.

As autonomous systems gain more autonomy, the uncertainty surrounding liability could translate into economic costs for businesses. Raut expects lawmakers to address this issue as the stakes become higher, potentially leading to clearer regulations. Additionally, the rise of agentic AI could impact competition, as dominant platforms may restrict transactions to their own agents, reinforcing existing digital ecosystems.

Businesses will need to grapple with questions of responsibility, control, and the potential concentration of power in the hands of these gatekeepers.