L.A.'s Digital Archives Are Riddled With Duplicate Images — and the Numbers Show Why That's a Problem

Los Angeles city departments are carrying an estimated 34 percent rate of duplicate image files across their shared digital asset systems, according to internal audits reviewed by data management contractors working with the city's Information Technology Agency. The redundancy is costing real money at a moment when every dollar in the municipal budget is spoken for.

The timing is not incidental. With the 2028 Summer Olympics now less than two years out, the city is pushing hard to digitize construction documentation, permitting records, and public-facing media libraries for venues from SoFi Stadium in Inglewood to the Intuit Dome in Hawthorne. Duplicate image files embedded in those systems slow search functions, inflate cloud-storage invoices, and — most critically — introduce version-control errors that can cause the wrong architectural rendering or outdated site photo to circulate among contractors and press offices.

The Scale of the Problem in L.A.'s Own Systems

The Los Angeles County Department of Arts and Culture manages a digital archive that, as of its most recent public-facing report from fiscal year 2024–25, contained more than 2.1 million image assets across programs tied to the Getty Center, Grand Park, and the city's public mural registry. Duplicate-image-replacement workflows — automated processes that scan archives, flag identical or near-identical files, and swap redundant copies for a single canonical version — have been piloted by at least two county-adjacent organizations, including the Los Angeles Public Library system, which operates 73 branch locations and has been migrating legacy photo collections to a centralized cloud platform since 2023.

The math on storage waste compounds quickly. Commercial cloud storage for uncompressed image archives runs roughly $0.023 per gigabyte per month on standard enterprise tiers. A library or city agency sitting on 50,000 duplicate high-resolution images — each averaging 8 megabytes — is burning approximately $110 a month on files that carry no additional informational value. Multiply that across a dozen city departments, and the annual figure climbs past $15,000 in pure redundancy cost before accounting for the staff hours spent manually managing mislabeled or duplicated assets.

At the Los Angeles Homeless Services Authority, which administers the city and county's coordinated homeless response programs, the challenge is more operationally urgent. LAHSA's outreach teams document encampment sites, shelter bed availability, and client intake records with photographs that feed into dashboards used by Mayor Karen Bass's Inside Safe initiative. When duplicate images populate those dashboards — a site photographed twice by different outreach workers on the same day, for instance — case managers can miscalculate available resources or overcount cleared encampments. LAHSA processed more than 22,000 unique shelter placements in calendar year 2024 under Inside Safe, according to city-published data, and even a small percentage of image-data errors in that volume represents hundreds of potentially mismatched records.

What Automated Deduplication Actually Fixes — and What It Doesn't

Duplicate-image-replacement tools use perceptual hashing algorithms to compare files not just by filename or metadata but by visual content, catching near-duplicates that differ only in resolution or slight cropping. The technology is not new — it has been standard in enterprise digital-asset-management platforms since at least 2018 — but adoption across public-sector agencies in L.A. has been uneven. The city's ITA issued updated digital-asset guidance in March 2026 as part of its broader IT modernization roadmap, but individual departments retain discretion over their own storage vendors and workflows.

For Olympic planners at LA28, the pressure to clean up digital archives before international media and logistics partners begin accessing shared asset portals is real and time-sensitive. The organization has said publicly that its venue documentation systems will need to support thousands of credentialed users by early 2027.

Organizations and agencies looking to get ahead of the problem have a practical starting point: most enterprise platforms — including those used by the L.A. County Metropolitan Transportation Authority and LAHSA — already include deduplication modules that can be activated without additional licensing costs. Running an initial audit, identifying the duplicate rate, and setting a remediation schedule before the next fiscal year's storage contracts renew in the fall is the step that data managers say is most consistently skipped. The numbers make the case for not skipping it again.