Singapore's War on Duplicate Images: What Officials, Experts and Key Figures Are Saying

Singapore's Infocomm Media Development Authority flagged the issue formally in a circular earlier this year: duplicate images — identical or near-identical visual assets replicated across government digital services, commercial listing platforms and archival databases — are consuming measurable storage bandwidth, distorting search results, and in some cases presenting outdated or misleading information to end users. The problem is not new. The urgency, experts say, is.

The timing matters because Singapore is midway through its Digital Government Blueprint refresh cycle, with agencies under pressure to demonstrate leaner, more accurate digital infrastructure ahead of the Smart Nation 2.0 framework targets set for 2027. Bloated image libraries slow page load times, inflate cloud hosting costs and, in the case of property and retail platforms, can mislead consumers who encounter superseded photos of units already sold or products no longer available.

Where the Problem Shows Up

PropertyGuru and the Housing Development Board's own HDB Flat Portal have both drawn scrutiny from digital practitioners for carrying replicated listing images — sometimes the same exterior photograph of a Tampines or Bukit Batok block appearing across dozens of separate resale listings with different asking prices and attributes. Neither platform has publicly quantified the scale of the duplication, but practitioners working in the real estate technology space describe it as systematic rather than incidental.

The National Library Board's digital archive programme, which houses millions of images spanning Singapore's history from Raffles Place redevelopment in the 1960s through to current urban transformation projects along the Jurong Lake District, has itself grappled with deduplication as collections digitised at different resolutions are merged. The NLB has not disclosed specific figures on duplicate assets removed during its ongoing migration work.

Govtech, the government technology agency based at Mapletree Business City in Pasir Panjang, has been developing hashing-based image deduplication tools as part of its broader data quality mandate under the Singapore Government Tech Stack. Technologists familiar with the stack describe perceptual hashing — a technique that identifies visually similar images even when file names or metadata differ — as the most practical solution at scale, though implementation across legacy systems remains a friction point.

What the Experts Are Saying

Academics at the School of Computing and Information Systems at Singapore Management University, located on Bras Basah Road, have in recent months published work on automated image provenance tracking for public databases, arguing that deduplication is inseparable from broader questions of data integrity and misinformation risk. Their position, laid out in research presented at a regional data governance conference in May 2026, is that duplicate images are not merely a storage inefficiency — they are a structural accuracy problem.

On the commercial side, logistics and retail operators using platforms integrated with Lazada Singapore and Shopee's local marketplace have noted that duplicate product images — often uploaded by multiple resellers for the same SKU — complicate automated quality checks and can push inaccurate product information to the top of recommendation algorithms. Platform operators have described ongoing deduplication as a continuous process rather than a one-time fix.

Cloud storage costs give the conversation a financial edge. Amazon Web Services S3 storage, widely used by Singapore-based platforms, runs at rates that make terabyte-scale image duplication a non-trivial line item in operational budgets. With Singapore's digital economy projected by the Ministry of Trade and Industry to contribute 17 percent of GDP by 2027, even incremental inefficiencies at infrastructure level attract attention from both boardrooms and government technology offices.

Practitioners advise organisations to begin with a content audit using open-source perceptual hashing libraries, establish a single-source-of-truth image repository before any platform migration, and set automated ingestion rules that reject duplicates at upload rather than cleaning up after the fact. For HDB flat sellers uploading photos to the resale portal, the practical advice is simpler: use current photographs, taken after the most recent renovation, and remove older images manually before a listing goes live. The back-end systems, officials suggest, will increasingly do the rest.