OnX Shares Blueprint for Enterprise AI Rollout That Hit ROI in Under Three Months

North American technology services company OnX has published the results of deploying enterprise AI across its 2,300-person workforce, reporting a return on investment in under three months with no major security incidents. The company used itself as its own first customer, a model it calls "Client Zero," before extending the approach to external clients through a commercial service called Forge AI.

Key Takeaways

• OnX deployed Claude Enterprise across three distinct workflows: Claude.ai for research and documentation, Claude Code for engineering and technical tasks, and Anthropic's Cowork platform for cross-functional collaboration.
• The deployment achieved ROI in under three months, with the company attributing the outcome to treating governance and cybersecurity as foundational requirements from the start rather than adding controls after rollout.
• CISO Chris DeBrunner described the core challenge for security leaders evaluating AI: how to capture AI capability without accepting unacceptable risk. The company's answer was to establish vendor evaluation, threat testing, and operational oversight before any AI agents entered production environments.

The "Client Zero" approach allowed OnX to evaluate AI benefits and limitations in a real operating environment, identifying where outputs required human review and where governance structures were needed before commercializing the framework. The report highlights the emergence of agentic AI, describing systems that undertake multi-step tasks with varying degrees of autonomy, as the next phase of enterprise deployment.

Read the full article on Digital Journal