The Sticky Tape Problem: Why AI Agents Need a New Corporate Structure

When electricity first arrived in manufacturing, early adopters simply replaced their massive steam engines with giant electric motors, keeping the old, convoluted system of belts and shafts intact. It wasn't until factories were completely redesigned—placing smaller motors directly on individual workstations—that the true productivity boom occurred.

Today, businesses are making a remarkably similar mistake with artificial intelligence. As highly capable AI agents enter the workforce, companies are merely slotting them into traditional human hierarchies. Prasun Shah, Chief AI Officer at PwC UK Consulting, aptly calls this the "sticky tape problem." Organizations are layering autonomous digital workers onto legacy human operating models, which is akin to taping over a breaking system rather than fixing it.

The ambition is certainly there: 85% of organizations want to become "agentic" within three years. Yet, 76% admit their current infrastructure and workflows are fundamentally unequipped for the shift.

To bridge this gap, enterprise AI platform Ema and HFS Research have introduced a new framework: Agentic Business Transformation (ABT). According to Ema CEO Surojit Chatterjee, ABT is entirely different from past technological shifts. If digital transformation was about turning paper into software, and the "co-pilot" era was about assisting humans, ABT is about weaving AI agents directly into the fabric of the organization as independent actors.

This transformation demands a total teardown of the industrial-era corporate structure across three pillars. First, the technology stack must evolve. Instead of treating AI as just another software layer, it must become the "connective tissue" that moves seamlessly across applications to execute complex, multi-step workflows. This architectural shift can slash the time it takes to launch a new business process from months to mere days.

Second, the workforce itself requires a radical redesign. The traditional corporate pyramid, designed to optimize human output through layers of middle management, is becoming obsolete. As AI agents take over execution and coordination, human managers will transition to leading hybrid teams. Their new challenges will involve navigating psychological safety, status dynamics, and human-AI trust. The scale of this shift is massive; McKinsey estimates that by 2030, three-quarters of current jobs will need to be redesigned or redeployed.

Finally, companies must rethink their success metrics, shifting their focus from raw output to actual business outcomes.

Ultimately, unlocking the 30% to 50% process acceleration promised by AI agents isn't a coding challenge—it is an organizational one. Buying the smartest AI is useless if you force it to work in a 20th-century bureaucracy.

Key Points

• While 85% of companies want to use autonomous AI agents, 76% lack the organizational readiness to do so.
• Layering AI agents onto legacy human workflows creates a 'sticky tape problem' that limits the technology's potential.
• Agentic Business Transformation (ABT) requires rebuilding the technology stack to use AI as connective tissue across systems.
• Management structures must adapt to lead hybrid human-AI teams, with McKinsey predicting 75% of jobs will need redesigning by 2030.

Why It Matters

The true bottleneck in enterprise AI adoption is no longer the technology itself, but the legacy organizational structures that constrain it. Companies that fail to redesign their workflows will miss out on massive productivity gains.

Sources:

• Rethinking organizational design in the age of agentic AI — MIT Technology Review - AI