Two Bangladeshi tech entrepreneurs have introduced an anonymous crowdsourced platform, dubbed ‘GhushSite’, designed to document informal financial demands across public sector offices. Built on the foundational principle of keeping personal identities strictly confidential, the initiative enables citizens to log instances of bribery without creating user accounts, registering email addresses, or disclosing telephone numbers.
The portal is the brainchild of Saif Kabir and Shahed Sharif, who developed the Bangladesh-specific platform in a single day after drawing inspiration from a comparable Indian civic-tech initiative. The website allows users to state the relevant public department, the service sought, the approximate location, the sum requested, and the ultimate outcome of the encounter. However, contributors are strictly barred from naming individual government officials or publishing personal identifying details.
Kabir highlighted that while widespread public awareness of bribery exists throughout the country, precise data regarding the exact amounts demanded has historically remained opaque. A citizen seeking a driving licence, passport renewal, or land mutation often learns of extra-legal “processing charges” without knowing whether the demand reflects a routine rate or extortionate inflation. The objective of the open ledger is to build a community-driven repository that exposes systematic transactional patterns over time.
Within roughly 48 hours of its launch, the portal logged 458 unverified user submissions across the country’s eight administrative divisions, reflecting aggregate demands of approximately 24.5 million Bangladeshi Taka. The geographic distribution of reports shows Dhaka leading with 254 submissions, followed by Chittagong with 98, Sylhet with 24, Rajshahi and Khulna with 22 each, Mymensingh with 16, Rangpur with 15, and Barishal recording seven reports.
Although the arithmetic mean across submissions calculated out to roughly 33,747 Taka, the creators noted that the figure was skewed upwards by several high-value allegations within customs, taxation, and land administration involving millions of Taka.
The platform operates through a automated three-step workflow. Once a submission is entered, it passes through an automated moderation filter powered by a natural language processing model. The system scans the text to strip out personal names, mobile numbers, National Identity (NID) details, file reference numbers, and abusive language before publishing. Kabir clarified that this automated mechanism functions solely as a privacy safeguard rather than a fact-checking instrument, meaning individual entries remain entirely unverified self-reports.
The creators emphasise that the project is not designed to replace statutory anti-corruption watchdog bodies or serve as a judicial grievance cell. Instead, they view it as an awareness ledger. While an isolated report of a 10,000 Taka demand for a municipal licence might easily be dismissed, forty identical submissions from the same branch reveal an undeniable operational trend. Funded privately by the founders and supported through voluntary user contributions, the platform aims to empower citizens, investigative journalists, and policy researchers by breaking the habitual silence surrounding informal extortion.
