State Customs Committee of Uzbekistan Uses IBM watsonx.ai and Llama 3.3 to Automate 80,000 Customs Declarations

Facing a rising volume of customs declarations with complex free-text goods descriptions, the State Customs Committee of the Republic of Uzbekistan, led by Chairman A. Yu. Mavlonov, worked with IBM Expert Labs and IBM Client Engineering to build an AI system for automated goods classification and declaration processing.
The system uses IBM watsonx.ai as the development studio and watsonx.data to process distributed data at scale, with a Llama 3.3 large language model extracting structured data from free-text declaration fields and classifying products against Uzbekistan's customs codes. According to IBM's published case study, the tool cut per-declaration processing time from 2-3 hours to seconds, and in 2024 more than 80,000 declarations worth a combined $5 billion were completed without staff involvement; the project also reclassified 44,000 commodity items across 93 customs codes into 12 additional categories, re-registered 80 declarations covering 975 items, and identified an extra $246,000 in revenue. Separately, the Committee's IT department has expanded its digital footprint from 9 information systems in 2021 to 37 in 2025, interconnected with 33 government agencies, and introduced 'AiRo', a passenger-facing customs-guidance robot at Tashkent International Airport intended to complement, not replace, human officers.
The performance figures come from a vendor-published case study, which carries a standard disclaimer that 'actual results will vary based on client configurations and conditions,' and no independently audited accuracy or error-rate assessment of the classification model has been made public.

Read the full case study: https://www.ibm.com/case-studies/state-customs-of-uzbekistan

Read the full analysis: https://www.ibm.com/case-studies/state-customs-of-uzbekistan