Wanted: Token spend control and governance without slowing down agentic AI innovation (www.techradar.com)

The rapid rise of agentic AI adoption in enterprises has highlighted significant concerns regarding governance and cost control, leading to potential risks such as data leakage and unforeseen expenses. As companies increasingly transition from basic AI functions to complex autonomous agents, the traditional governance frameworks struggle to keep pace, drawing parallels to earlier challenges seen during the advent of cloud computing. With predictions that over 40% of agentic AI projects might be scrapped by 2027 due to these cost and risk management issues, it has become critical for organizations to find solutions that allow innovation without jeopardizing financial and operational integrity. To address these challenges, there is a growing call for a control plane or AI gateway that offers a centralized oversight mechanism for AI operations. This would enable organizations to manage token consumption effectively and maintain compliance, whether AI models are hosted in public clouds or on-premises. By implementing this control system, CIOs can better navigate the complexities of agentic AI while promoting efficiency and maintaining their role as catalysts for change, ensuring that the transformative potential of AI is harnessed responsibly and sustainably.