Navigating the Governance Challenges of Agentic AI

The rise of agentic artificial intelligence (AI) — systems that operate autonomously with minimal human oversight — presents both unprecedented opportunities and significant governance challenges, according to experts at the Special Competitive Studies Project (SCSP), a nonprofit and nonpartisan initiative focused on strengthening America's long-term competitiveness in AI. Unlike current AI models that respond to prompts, agentic AI can independently set goals, create plans, and execute multi-step tasks, potentially creating a self-accelerating loop where AI helps build better AI.

Ylli Bajraktari, president of SCSP, warned in a recent newsletter that this compounding capability development could far outrun current projections. He emphasized the security implications: "An agent that can navigate complex bureaucratic systems, identify exploitable vulnerabilities, and act without leaving a clear attribution trail represents a qualitative expansion of adversarial capability." The United States must recognize that adversaries will deploy agentic AI in areas where governance is weakest, using it for coercion, espionage, and influence.

Effective governance, SCSP experts explain, does not focus solely on the AI model itself but on the "scaffolding" built around it. This scaffolding includes connectors to bridge the model to real-world infrastructure such as email, booking systems, and financial platforms; memory that allows the system to learn and adapt over time; planning capabilities to break objectives into smaller tasks; permission structures defining system access; and guardrails determining what the system will refuse to do, such as spending limits or human sign-offs.

Accountability remains a critical challenge, with governance falling short in three key areas. First, responsibility is untraceable when AI acts on behalf of a user — there is no way to determine who authorized what. Second, current frameworks evaluate whether a task was completed, not whether it was performed safely or caused harm. Third, agentic AI builds personal profiles that may include sensitive data inferred from patterns of behavior, preferences, and actions, often without explicit consent.

Despite these challenges, SCSP experts stress that agentic AI is not a technology to be feared. Institutions that prioritize understanding, shaping, and governing agentic AI will determine their own competitive position and influence the global environment in which these systems operate. For more insights, visit scsp.ai to learn how the United States can pursue effective governance of agentic AI.