evidoria

← Back to browse

Good practice Imported

AI-Assisted Timber-Fraud Detection on Romania's SUMAL 2.0 Wood-Tracking System

Romania · Bucharest · See the Romania profile

Romania's Environment Ministry piloted Google Vertex AI on its SUMAL 2.0 timber-permit system, scanning 400,000 monthly permits and 1.6M photos in three days, triggering 7 criminal probes and $930,000+ in fines — though it still misses subtler photo fraud, OCCRP found.

AI-Assisted Timber-Fraud Detection on Romania's SUMAL 2.0 Wood-Tracking System

Details

Promoter
Ministry of Environment, Waters and Forests (Romania), with Google Cloud and Zitec
Period
System live since 2021; AI pilot November 2023, ongoing
Keywords
forestry, environment, anti-fraud, public administration

Description

Romania is one of the EU's most heavily logged countries, with the government's own estimates suggesting the annual illegal harvest volume is roughly double the legal quota. Since January 2021, the Ministry of Environment, Waters and Forests has required every timber transport to be logged in SUMAL 2.0, a national digital tracking system: drivers must photograph loads and upload permits in real time, cross-checked against satellite forest-change alerts refreshed every 2–7 days.
Investigative reporting by OCCRP found haulers were re-using or “cloning” old photos across different loads to fake compliance. In November 2023 the Ministry, working with Google Cloud and Romanian IT firm Zitec, piloted Vertex AI to automatically scan a full month's permits and photos — roughly 400,000 permits and 1.6 million images — a task completed in about three days (0.64 seconds per permit) versus an estimated 125 people needed for manual review, against a Forest Guard staff of about 200. The pilot flagged enough suspicious cases to trigger seven criminal investigations and fines/confiscations of wood worth more than $930,000. Over a broader roughly two-year enforcement window, the wider SUMAL monitoring regime is credited with uncovering more than 2,000 fraudulent “clone” transports, €2 million in fines and €1.6 million in confiscated wood.
OCCRP's investigation found the AI reliably caught blatant fakes (photos of chairs, faces, tires) but “wasn't sophisticated enough to detect more subtle frauds, like images of real wood” reused across different loads, so plausible clone photos could still slip through. The Ministry has since pursued a follow-on €8.9 million tender, partly funded through Romania's EU recovery plan (PNRR), for license-plate recognition cameras and LiDAR to supplement photo review — treating the AI pilot as a first-generation tool rather than a finished solution.

Read the full analysis: https://www.romania-insider.com/romania-ai-check-timber-transports-apr-2024

Implementation

Implementation detail (cost, timeline, staffing, conditions for success) is not yet available for this practice.

Do you run this practice? Claim it — verified implementers get a public contact pathway and can propose corrections.

Data sources

Where this practice's information was retrieved from, and when.

Attachments

Similar practices you may find useful