Which of the following must you do before connecting to a dataset in order to train a machine learning model in SAP Al Core? Note: There are 2 correct answers to this question.

Question

Options

• AStore the dataset in a hyperscaler object store.
• BGrant access rights to the SAP BTP cockpit.
• CProvide the storage secret to access the dataset.
• DStore the dataset in the SAP HANA Vector Engine.

• A
  75% (21)
• B
  7% (2)
• D
  18% (5)

Explanation

Before SAP AI Core can train a model, it needs two things: somewhere to read the data from, and permission to access it. Option A is correct because SAP AI Core integrates with hyperscaler object stores (like AWS S3, Azure Blob, or Google Cloud Storage) as its data source - it does not natively host datasets itself. Option C is correct because you must register a storage secret (credentials) in AI Core so it can authenticate and pull data from that object store at training time.

Option B is wrong because SAP BTP cockpit access rights govern the BTP platform UI, not dataset connectivity for AI Core training workflows. Option D is wrong because SAP HANA Vector Engine is a database feature used for vector/embedding search, not the storage target AI Core uses to read training data.

Memory tip: Think of it as two-step access - first park the data (hyperscaler object store = A), then hand over the keys (storage secret = C). No parking spot + no keys = no training.

No community discussion yet for this question.