Innovation Management Watch Summary: “State of Generative AI in the Enterprise” by Deloitte

This week’s Innovation Management Watch Summary features Deloitte’s State of Generative AI in the Enterprise report—an in-depth look at how organizations are progressing from experimentation to scale in one of the fastest-evolving technology shifts of the decade. 

The study reflects the voices of thousands of executives across industries and regions. It captures a critical moment: as the initial excitement around generative AI gives way to the harder work of building sustainable business value, leaders are confronting a reality check. The promise remains enormous—but the path to impact is uneven. 

Deloitte finds that while enthusiasm for GenAI remains high, most companies are still struggling to turn potential into measurable results. Only a small share of organizations have reached the point of scaling applications across departments or embedding them into workflows. The majority remain stuck in pilot mode, held back by governance gaps, regulatory uncertainty, and inconsistent data readiness. Progress, the report concludes, is happening at “the speed of organizational change, not the speed of technology.” 

Key Findings 

ROI remains elusive. 

Only 74% of organizations report that their most advanced generative AI initiatives meet or exceed ROI expectations—and those successes are concentrated among early adopters with strong governance and cross-functional collaboration. 

Certain functions are pulling ahead. 

Cybersecurity, IT, and operations lead in realizing tangible value, where GenAI is improving detection, automating tasks, and accelerating troubleshooting. By contrast, functions like sales, marketing, and R&D lag behind, often due to unclear use cases and fragmented data systems. 

Limited access, limited adoption. 

Fewer than 40% of employees have regular access to GenAI tools. Most organizations have yet to establish enterprise-wide frameworks for responsible and equitable AI use, leading to silos and “shadow AI” experimentation outside governance boundaries. 

Barriers persist at the structural level. 

The most common obstacles include regulatory uncertainty, the absence of clear governance, inconsistent risk management, and lack of clarity around data ownership. In heavily regulated sectors—such as finance and healthcare—compliance concerns have slowed progress more than technical limitations. 

Agentic AI is the next frontier. 

A notable trend emerging in Q4 2025 is the rise of agentic AI—autonomous, goal-driven systems capable of performing complex tasks and collaborating across enterprise applications. These tools are moving from theoretical promise to early adoption, signaling the next evolution in enterprise AI maturity. 

The Leadership Imperative 

Deloitte’s central message is clear: technology alone does not create transformation—organizations do. C-suite optimism remains high, but many leaders are overestimating their organization’s readiness to scale. 

To close the “execution gap,” executives must evolve from AI cheerleaders to AI champions—shifting from promoting experimentation to driving disciplined, strategic deployment. That means aligning governance, risk, and workforce readiness under a unified vision. It also means ensuring trust and transparency underpin every initiative, from model training to employee enablement. 

High-performing organizations distinguish themselves by three core behaviors: they invest early in reskilling, they integrate GenAI into existing digital strategies rather than treating it as an add-on, and they adopt agile governance models that allow innovation while managing risk. 

Deloitte’s research underscores that scaling GenAI isn’t about chasing the newest tool—it’s about operationalizing trust, human capability, and responsible design. Those who get this right will turn experimentation into enterprise advantage, positioning their organizations to thrive amid the next wave of intelligent automation. 

At a Glance 

• Only 1 in 4 companies are achieving measurable GenAI ROI. 

• Governance, regulation, and data readiness remain the top barriers. 

• Agentic AI and multi-agent systems are emerging as enterprise focus areas. 

• Success depends on leadership maturity, not technological novelty. 

This summary is based on Deloitte’s report “State of Generative AI in the Enterprise” All rights to the original content remain with the respective copyright holders.