This guide is to help users navigate the new Kpler AI prototype available on the mobile Kpler app in order to get the most out of it.
How can Kpler AI prototype on the mobile app help me?
Kpler AI on the mobile app can help you in two different ways, thanks to two sources of data:
Analytics without interpretation or analysis:
You can ask for specific quantitative information about commodities and fertilizers, including: flows (imports/exports), inventories, supply and demand, traders (cargo-by-cargo details about loadings/discharges, vessel locations), refineries (refinery operations), vessel congestion, and fleet metrics (floating storage or commodity on water).
Example question that will source Kpler analytics data:
What are the exports of crude oil from Saudi Arabia to China in March 2023?
Insights:
Expert-driven commodity research and news with daily, weekly, and monthly analysis to deliver actionable trading and market intelligence. You can ask question types that require a thorough understanding of the commodity market, request a detailed market analysis or interpretation, or involve geopolitical events linked to the commodity market.
Example question that will source Kpler insight data:
What is the situation in the Red Sea?
What is included in the Kpler AI prototype on the mobile app?
Depending on your permissions and the products you are interested in, you can ask Kpler AI questions on insights, flows (imports/exports), inventories, supply and demand, traders (cargo-by-cargo details about loadings/discharges, vessel locations), refineries (refinery operations), vessel congestion, and fleet metrics (floating storage or commodity on water).
Kpler AI does not yet handle queries on compliance, power, or Kpler methodologies.
How to use the Kpler AI prototype on the mobile app properly?
To get the most accurate responses, include specifics like country, region, time period, and product.
How does the Kpler AI prototype on the mobile app work?
It analyzes the question to extract Kpler-specific metadata (installations, product, vessels, zones, timeframe). Then, based on the query, it creates a corpus of:
Insight documents
Quantitative data from Kpler APIs
Finally, it generates an answer to the question based on the corpus it has built up.
How do I know if the answers are correct?
When the answer is generated, you can access the sources of this answer by clicking on the “sources” button. You can find the article written by our experts from which the answer comes or the API for quantitative data.
If you have any feedback, please contact us at this email address: product@kpler.com or reach out to your Customer Success Manager.