Advancing the War Data Platform

Data is the new ammunition. Protecting it, understanding it, accessing it, and using it is key to intelligence awareness, battlefield success and management effectiveness. In addition, as the Pentagon announced the clean audit of its War Data Platform (WDP), Deputy Chief Financial Officer Tom Harker said, “Reliable data is the bedrock of financial accountability.” The Pentagon made important moves over the last year to harness the power of data. One element of the data strategy is the referenced WDP – formerly called Advana – which leverages artificial intelligence for both business systems management and warfighting outcomes. While this system undergoes evolutionary changes and endures acquisition strategy confusion, clarity on its status and way ahead is necessary. Advana was created by the comptroller to become the Pentagon single source of truth for data. Starting with financial systems to help the military move toward a clean full financial statement audit opinion, it is the foundation of a systemic and cultural shift toward something much more – comprehensive, timely, reliable data supremacy for successful management and operational outcomes. A positive audit outcome, though legally required and beneficial, is not the real goal. The reason for the audit is ultimately to improve data quality, ask the right questions of data, and therefore support decision-making that creates value for the warfighter. When Advana moved from the comptroller to the Chief Digital and Artificial Intelligence Office (CDAO) due to the key role it plays in leveraging artificial intelligence, it continued to evolve, adapt and expand, but it also lost some of the independent, expert management it needed to achieve its ultimate vision. There are three key things to know to understand the current situation and what should happen next. First, the WDP/Advana remains central to achieving a clean audit for the department in 2028 and delivering enterprise data to support operational missions. Second, the Pentagon, through CDAO, has signaled a shift in priority by separating Advana into two components: WDP (focused on expanding the core data layer and minimizing one-off tools and solutions) and Advana for financial management (focused on audit remediation and follow on activities to understand financial posture). Third, evidenced by a confused – and at times contradictory – acquisition strategy, government leadership in CDAO is bifurcated and lacks a single technical and mission-oriented champion setting a vision and multiyear strategy to focus the value of WDP and orient customer expectations. CDAO recently awarded an $820 million labor-based, cost-plus contract to Accenture Federal Services with a largely unchanged scope from previous years. CDAO also awarded significant contracts to Databricks, Amazon Web Services, and Palantir to support WDP. The underlying technology available to CDAO has changed in recent years offering more flexible management options that can be maintained and enhanced with fewer technical staff, calling into question the approach on these contracts. Both actions are also seemingly at odds with previous White House and Pentagon guidance and put the future of the WDP on shaky ground. With these points in mind, there are three actions that could move the capability ahead. First, move overall management the WDP and Advana back to the Comptroller, who not only should own the requirements for managing the end-to-end control environment for attestation to future auditors, but also has more expertise and mission focus on audit and can provide the guidance and oversight necessary for the mission specific AI incubation while leveraging acquisition expertise of multiple partners. Since the audit is still intimately intertwined with WDP using essentially the same infrastructure backbone, it makes sense to maintain a single source of truth. Second, return more access, ownership, and funding to customers to own the future requirements for maintaining the data, tools, policies, and infrastructure for their own missions. CDAO has promised multi-tenant development tools and a move to entirely self-service with deeper access for its customers. Those capabilities have largely been available for months while additional required tools and support have not been provided. Customers should have this access and control. Third, shrink reliance on a heavy central WDP team that adds bureaucracy to simple technical and data policy decisions and explore more fixed price models that drive true outcomes like data reliability, availability, resilience or standardization – which are the right outcomes of a defense data fabric. At present acquisition specialists are driving technical requirements but not focusing on aligning with acquisition guidance. Technical specialists are not empowered. Strategic vision is lacking. The result is an ill-constructed acquisition that is struggling to drive innovation and a strategy that supports the customer base. As the Pentagon pursues an overall full financial statement audit and the fundamental institutional improvements to data management decision-makers and warfighters need, the management and acquisition approaches to data must align and keep pace.