As enterprises transition from debating artificial intelligence adoption to actively deploying it, AI governance has become a more complex challenge. Autonomous agents now operate continuously across various platforms, and traditional controls designed for deterministic software are proving inadequate, as first reported by Silicon Angle.
At Dell Technologies' AI Leadership Symposium, cost and control were central themes in discussions about enterprise AI deployment. Shiv Agarwal, co-founder and CEO of Singulr AI, highlighted the difficulty in rationalizing overlapping controls and ensuring their effectiveness in practice. He emphasized the need to bridge the gap between enterprise intent and the actual runtime state of AI agents, noting that governance is becoming crucial yet increasingly challenging due to the rapid pace of AI adoption. Experts also discussed the importance of assessing agent behavior during execution, with some customers opting to restrict agent operations within their virtual private clouds to maintain data control. The need for standardization across AI governance frameworks, such as the European Union's AI Act, and the necessity for enterprises to keep pace with rapid AI development were also key discussion points.
Source: Silicon Angle
