Sama introduces Bulk Annotation to increase AI labeling by 80%

Sama has introduced Bulk Annotation, a new feature aimed at tackling repetitive work in AI data labeling processes. The company’s latest advancement allows groups of nearly identical items to be annotated together rather than individually, eliminating persistent inefficiencies for companies processing large training datasets.

Operational implications

Repeatedly labeling thousands of similar or duplicate items is a well-known challenge across industries where AI is used. Manual annotations consume significant resources and can result in inconsistent datasets. According to the company, initial pilots of Samas Bulk Annotation indicate up to 80% throughput improvements and up to 25% reduction in annotation inconsistencies.

Platform approach

The Bulk Annotation system uses embedded machine learning capabilities to group duplicates, variants, and near matches within a data set. Annotators can then review and classify these groups in a single action,…