Univé AI Workforce Transformation

Univé, one of the Netherlands’ largest cooperative insurers, has transitioned from treating AI as a tool deployment to an organizational transformation by deploying ChatGPT Enterprise. This strategy focuses on building AI capability across the entire workforce, enabling employees to redesign their own workflows and solve internal challenges independently.

Scaling AI Through Employee Capability

Univé has shifted its focus from building centralized AI solutions to creating a workforce of "builders." By providing employees with the permission, structure, and dedicated time to rethink their work, the organization has moved away from requiring detailed business cases for every new AI application.

Key adoption metrics include:

  • License Activation: 97% of ChatGPT Enterprise licenses have been activated.
  • Weekly Engagement: 85% of licensed users are active every week.
  • Usage Volume: Active users average 40 prompts per week.
  • Custom Tooling: Approximately 1,500 custom GPTs have been created by employees to solve internal operational challenges.

Governance as an Innovation Accelerator

Univé integrated governance into the rollout from the first day to ensure employee trust and security. Rather than acting as a barrier, these guardrails serve as an accelerator for responsible experimentation.

The governance framework includes:

  • Security and Privacy: Enterprise authentication, privacy assessments, and security reviews.
  • Access Control: Connector permission inheritance ensures AI access is limited to what employees are already authorized to see within enterprise systems.
  • Accountability: Clear human accountability and responsible AI principles are maintained, ensuring that while AI prepares work, humans make the final decisions.

High-Value AI Applications in Insurance

AI is currently utilized across virtually every knowledge-based function, including finance, HR, legal, IT, and customer service, with specific high-impact applications in claims and underwriting.

Pet Insurance Claims

Workspace Agents are used to assemble claim files, review veterinary invoices, check policy conditions, and identify anomalies. This process reduces the preparation time for claims from hours to minutes, allowing claims professionals to focus on the application of their expertise while remaining fully accountable for the final decision.

Underwriting

Workspace Agents proactively review incoming work queues and combine information from approved enterprise sources. By flagging risk indicators and identifying missing documentation before an underwriter begins their day, the system allows underwriters to spend less time gathering information and more time making high-quality decisions.

Future Direction: Agentic Workflows

Univé is exploring the next phase of enterprise AI through the implementation of agentic workflows. These workflows are designed to proactively prepare recurring work across approved enterprise systems, surfacing relevant context and creating evidence-based starting points before an employee begins their task.

Strategic Leadership Lessons

Univé's approach provides several key takeaways for organizational AI adoption:

  • Organizational Capability over IT Implementation: AI should be treated as a fundamental change in how work is done rather than a software rollout.
  • Leadership Investment: Transformation requires dedicated AI leadership sessions to challenge managers to rethink the role of work.
  • Governance as a Foundation: Designing governance into the rollout from day one provides the confidence necessary for employees to experiment.
  • Employee Empowerment: Success is measured by sustained adoption and employee capability rather than simple productivity metrics.

Sources