Exploring AI and its uses in customs practice

In this recording, Dr Ramūnas Šablinskas, IT solution architect and technical product manager at CC Learning UAB (CustomsClear), explains how artificial intelligence (AI) works and how it can be used to automate routine customs tasks, giving customs and trade compliance professionals a practical grounding for evaluating AI tools in their own organisations.

Topics covered:

• What AI is and why large language models (LLMs) have suddenly made it mainstream
• The data, information, knowledge and wisdom hierarchy, and why machines lack true "knowledge"
• How AI recognises, classifies and predicts patterns, and how this relates to generative AI
• How AI models are trained, and the scale of parameters and compute involved
• Why AI models "hallucinate" and cannot be fully trusted to memorise exact facts, laws or regulations
• The particular risk this poses in customs, where regulations, tariffs and quotas change frequently
• How combining AI with curated, up-to-date knowledge libraries eliminates hallucinations and produces reliable, grounded answers
• How conversational AI assistants (such as Picoco AI) are built from an organisation's own expert knowledge and question-answer history, and how quickly a pilot project can be set up

For a broader overview of the topic, please watch the full recording. The slides are available in the Resources section.

Please note that this summary was generated using AI, based on the recording and available slides.