Unlike most countries with national police, the US has thousands of local agencies that historically could not share information effectively. This fragmentation is a major weakness that criminals exploit, creating a large opportunity for tech platforms that facilitate inter-agency data sharing.

To overcome data silos in a regulated environment, CIBC prioritized building internal trust. They proactively brought legal, compliance, and privacy teams together, clearly defining the use case and value of unified data, which was critical for gaining enterprise-wide approval.

A critical hurdle for enterprise AI is managing context and permissions. Just as people silo work friends from personal friends, AI systems must prevent sensitive information from one context (e.g., CEO chats) from leaking into another (e.g., company-wide queries). This complex data siloing is a core, unsolved product problem.

AI can easily write code for system integrations, but the primary bottleneck isn't coding—it's context. The real work involves tracking down employees to understand what ambiguous, legacy data fields actually mean, a fundamentally human task of institutional knowledge discovery.

The NCIC, a key FBI database for warrants and stolen vehicles, is more like a daily CSV file than a real-time system. This lag, combined with a lack of data integrity protocols, means outdated information, like a recovered rental car still listed as stolen, persists and puts civilians at risk.

Mass surveillance capabilities weren't created by a single administration. They are the result of decades of incremental, bipartisan decisions from Reagan to Obama, driven by political fears of appearing weak on national security, making the system deeply entrenched and difficult to reform.

The data infrastructure for law enforcement is fragmented and archaic. Until recently, some major US cities ran on paper, and states even outlawed cloud storage. This creates massive data silos that hinder investigations, as criminal activity crosses jurisdictions that don't share data.

Large enterprises inevitably suffer from "data sprawl," where data is scattered across on-prem clusters, multiple cloud providers, and legacy systems. This is not a temporary problem but an eventual state, necessitating tools that provide a unified view rather than forcing painful consolidation.

The urgent need to calculate exposure to Lehman during the 2008 crisis forced Goldman Sachs to centralize its disparate data. This crisis-driven project revealed the immense business value of data, shifting its perception from "business exhaust" to a strategic enabler for the firm.