What happens when AI moves from assisting to acting autonomously

Artificial intelligence (AI) is increasingly being adopted by governments worldwide to streamline service delivery and enhance public services. However, the emergence of agentic AI, which can autonomously perform tasks within defined parameters, has sparked discussions on the appropriate level of responsibility assigned to these advanced AI systems. The transition from AI providing assistance to AI acting independently raises questions about the extent of delegation and accountability for AI-driven decisions.

Unlike traditional AI, agentic AI possesses the capability to interact with digital tools, access information, and execute a series of actions to accomplish specific objectives. This advanced form of AI can streamline processes, reduce workload, and improve service delivery efficiency. For instance, Kisan e-Mitra, a voice-enabled AI chatbot in India, assists farmers by providing information on government agricultural schemes and addressing their queries, thereby simplifying access to services.

As governments evaluate the integration of agentic AI, they must consider the implications of delegating administrative tasks to AI systems. While this may lead to increased efficiency and reduced manual intervention, it also necessitates a reassessment of the boundaries governing AI autonomy. The "blast radius" of AI actions, encompassing the potential consequences of AI-driven decisions, becomes a critical concern.

Decisions made by AI agents, ranging from document verification to emergency response coordination, could have far-reaching implications, especially in a diverse nation like India where multilingual support is crucial.

Countries such as Singapore, the UAE, and the United Kingdom are at the forefront of exploring agentic AI governance frameworks. These initiatives aim to balance the benefits of AI-driven automation with the need for safeguarding against potential risks. The appointment of dedicated AI ministers in various nations reflects a growing recognition of AI's evolving role in governance, extending beyond mere technology adoption to encompass strategic planning and decision-making.

The challenge lies in determining the optimal balance between leveraging AI's capabilities to enhance public services and ensuring adequate oversight mechanisms. As governments contemplate the institutionalization of agentic AI, careful consideration must be given to how AI systems can operate effectively while remaining within acceptable ethical and legal parameters.

The successful implementation of agentic AI hinges on striking a delicate equilibrium between innovation and responsible governance, ensuring that the transformative potential of AI is harnessed for the betterment of society without compromising accountability or public trust.