By the Numbers: How Duplicate Images Are Quietly Draining LA's City Digital Archives

Los Angeles city agencies collectively hold tens of millions of digital image files — and a significant share of them are exact or near-exact duplicates, according to a review of internal IT procurement documents filed with the City Administrative Office this spring. The problem is neither glamorous nor politically charged, but the dollar figures attached to it are drawing new attention from department budget managers ahead of the fiscal year 2026-27 cycle, which took effect July 1.

The timing matters. Mayor Karen Bass's housing emergency declaration, now in its third year, has pushed the city's Planning Department and the Los Angeles Homeless Services Authority to digitize thousands of property inspection photos, site assessments, and shelter intake records at an accelerated pace. That volume surge — combined with a lack of deduplication protocols — has created a sprawl of redundant image data sitting across servers operated by the city's Information Technology Agency, which manages infrastructure for more than 40 departments.

The Storage Math Nobody Wants to Do

Cloud and on-premise storage is not free. Enterprise-grade storage contracts used by large municipal governments in the United States typically run between $0.02 and $0.08 per gigabyte per month under standard tiered pricing from vendors including Microsoft Azure and Amazon Web Services. For a city the size of Los Angeles, which the ITA has publicly described as managing petabyte-scale data environments, even a modest 10 percent redundancy rate across image archives translates to a recurring cost that compounds annually. No single public figure has been released for the LA-specific total, but city council budget testimony from March 2026 referenced a line item of roughly $4.1 million annually for ITA cloud storage across all file types — images among the largest contributors by raw volume.

The Los Angeles Department of Building and Safety, headquartered on Figueroa Street downtown, processes thousands of inspection records weekly. Staff photographers and field inspectors often upload the same site photo multiple times — once from a mobile device, once from email, sometimes again via a shared drive sync — without any automated flag stopping the duplication. The same pattern appears at the Bureau of Engineering, which manages infrastructure photography for projects stretching from the Sepulveda Basin Recreation Area in the west San Fernando Valley to the LA River revitalization corridor running through Elysian Valley and Frogtown.

The Los Angeles County Metropolitan Transportation Authority, while a separate entity from the city, faces an analogous problem on a larger scale. Metro's capital projects office documented in a 2025 annual report that its construction photo archive for the Purple Line Extension alone had grown past 800,000 individual files. Industry benchmarks from digital asset management firms suggest 20 to 30 percent of construction project image libraries contain duplicates or near-duplicates when deduplication tools are not deployed at the point of upload.

What the City Is — and Isn't — Doing

The ITA issued a Request for Information in April 2026 seeking vendors capable of automated image deduplication at scale, with responses due by May 30. That RFI, posted to the city's online procurement portal, specified a requirement to handle files stored in both JPEG and RAW formats and to integrate with the existing ServiceNow platform used across city departments. Whether a formal Request for Proposals has followed is not yet public record.

For residents and journalists filing California Public Records Act requests — which in 2025 numbered more than 22,000 citywide according to the City Clerk's office — duplicate image files directly slow fulfillment. Every redundant file a staffer must manually review to determine relevance adds time to a process that city departments are already under court-monitored pressure to accelerate.

The practical path forward for city IT managers involves deploying perceptual hashing algorithms — technology that identifies visually similar images even when file names or metadata differ — before new uploads complete. Several other large American cities, including Chicago and Houston, have implemented such systems at the departmental level. Los Angeles has the ITA infrastructure to do the same; the question now is budget allocation and political priority in a fiscal year already stretched by wildfire recovery costs and Olympic construction overhead.