Three AI Demos, Three Different PIM Jobs

Three AI demos on 16 November 2023 were three different jobs. Ergonode talked about product data and workflows inside the PIM. Connecting the Dots showed enrichment and product recommendations. Descrb showed generated product descriptions. Match the demo to the job that is slow in your catalog before you treat "AI for PIM" as one purchase.

The session was online at 15:00 CET. The recording:

Speakers:

• Tomas Hesseling of Ergonode discussed AI for product data management and for workflows.
• Arnout Schutte of Connecting the Dots showed AI for data enrichment and product recommendations.
• Sylvain de Backer with Michal Kulaczkowski of Descrb showed an AI platform that generates product descriptions.

The LinkedIn URL on Michal's name is a search page, not a profile, and the search keyword is cut off at "Micha". Both names stay as published.

A generated description still has to match an attribute you are allowed to publish. Enrichment and recommendations fail the same way when the target field has no owner. Product data quality and governance is who owns that field. AI-ready product data says what the record has to contain before a model writes or recommends from it. A later Onedot session, AI mapping still needs an owner for the target model, makes the same split on supplier files. The three demos are on the recording.