As organizations increasingly adopt enterprise-level AI agents, a notable gap is emerging between ambition and practical implementation. While a significant number of companies express a desire to integrate agentic AI into their operations within the next three years, many report that their current infrastructures and workflows are ill-equipped for such a transformation. This disconnect often stems from a lack of preparedness across various dimensions, including workforce processes and operational frameworks.
Prasun Shah, Global CTO for workforce consulting and Chief AI Officer at PwC UK Consulting, highlights the prevalent issue of layering AI agents onto existing operational models rather than fundamentally rethinking these models. He likens this approach to applying temporary fixes, such as sticky tape, to a system that requires comprehensive redesign. The inherent value of agentic AI lies in its potential to autonomously execute entire workflows, making decisions and adapting to changing conditions with minimal human intervention. In sectors like customer service and sales, early implementations suggest that AI agents could expedite business processes significantly, reducing low-value work time considerably. Yet, reaping these benefits necessitates an enterprise-wide shift in thinking and operations.
To facilitate this transformation, the term ‘agentic business transformation’ (ABT) has been introduced by the enterprise AI platform Ema in partnership with HFS Research. This concept aims to fill the void in existing terminology surrounding AI agents, providing organizations with a framework for understanding and adopting this technology. Ema’s CEO, Surojit Chatterjee, emphasizes that ABT transcends previous definitions of digital transformation, as it involves integrating AI agents deeply into organizational structures rather than merely enhancing existing processes with AI tools. ABT encompasses three core elements: the technology stack, the workforce, and success metrics. As AI agents become integral to organizational workflows, companies need to rethink their technology infrastructures to support the rapid processing capabilities of these agents, thereby enabling them to make informed decisions across multiple systems. This architectural evolution will not only enhance decision-making but also improve organizational adaptability, allowing companies to respond swiftly to new business requirements without waiting for lengthy software updates.
Source: Rethinking organizational design in the age of agentic AI via MIT Technology Review