Issue

StoneTurn was engaged by a multinational industrial corporation to investigate a competitor for alleged bid rigging. During the investigation, StoneTurn uncovered indicators that the competitor was continuing to conduct business with sanctioned Russian entities through intermediaries across the EU, Central Asia, and the UAE, potentially evading international sanctions.

To validate these concerns, StoneTurn needed to analyze highly fragmented global trade data spanning multiple jurisdictions, entities, languages, and formats—much of which was considered difficult or inaccessible to obtain. Critically, key shipping records were in Russian and described goods in dense, industry-specific technical language, while the controlled-goods lists they had to be tested against were maintained in equally specialized English.

Solution

StoneTurn deployed a multidisciplinary team that combined artificial intelligence, advanced data analytics, and traditional investigative tradecraft to map the network and surface hidden relationships.

• Global Trade Data Aggregation: Collected shipping and import/export records across dozens of countries, ingesting large-scale structured and unstructured datasets—in multiple languages and formats—into a single analytical framework.
• AI-Powered Translation and Goods Classification: Russian-language bills of lading described shipped items in technical jargon that defied simple keyword search. StoneTurn used AI to translate and interpret these descriptions and then match them—by meaning rather than literal wording—against the U.S. Commerce Control List (CCL), itself written in specialized English. This cross-language, meaning-based matching flagged controlled and prohibited goods that conventional text search would have missed entirely.
• AI-Driven Entity Resolution: Shell corporations, intermediaries, and ports appeared under inconsistent names, spellings, and transliterations across jurisdictions. StoneTurn applied AI to recognize when differently-written records referred to the same real-world entity, consolidating fragmented data into a coherent picture of ownership and transaction relationships.
• Network Mapping and Shipment-Flow Tracing: The AI-prepared and resolved data was assembled into a network model—entities, shipments, and routes represented as connected nodes—allowing the team to trace the flow of goods step by step through intermediaries, ports, and shell entities, and to expose pathways that were invisible in the raw records.
• Expanded Data Access: Sourced and analyzed import/export data from regions where reliable data was previously considered unavailable, materially improving the completeness of the investigative dataset.
• Integrated Intelligence Gathering: Combined open-source intelligence, discreet source inquiries, and AI-driven data analysis to corroborate findings and strengthen evidentiary support.

Results

StoneTurn uncovered a sophisticated sanctions evasion network that facilitated the flow of prohibited goods to sanctioned Russian entities through layered intermediaries and global trade routes.

• Translated, classified, and resolved trade data that powered a network model used to trace tens of thousands of shipments across multiple jurisdictions and shell entities.
• Identified controlled and prohibited goods by matching Russian technical descriptions against U.S. control lists across the language barrier—a method recognized as novel even among enforcement agencies.
• Revealed systematic efforts to circumvent U.S. and international sanctions frameworks.
• Delivered detailed reports and briefings to the U.S. Department of the Treasury (OFAC) and Department of Commerce (BIS), as well as European counterparts.

By combining AI, advanced analytics, and investigative expertise, StoneTurn transformed highly complex, fragmented, multilingual trade data into actionable intelligence—uncovering hidden misconduct at scale and demonstrating new methodologies for identifying sanctions evasion in an increasingly complex global trade environment.