Appen

Appen is a data services company that supplies human-labeled training data and annotation work for machine learning systems. It runs a global crowd of contributors, historically over a million people across more than 170 countries, who handle image labeling, audio transcription, text annotation, and relevance rating. The company has been in this business since well before the current AI boom, originally building its name in speech and language data collection before expanding into the broader data annotation market that now serves computer vision, NLP, and generative AI teams.

Its customers are mostly large AI labs and enterprises that need training or evaluation data at scale, particularly when the data has to span many languages, dialects, or cultural contexts that an in-house team can't easily cover. Appen offers both a managed service, where it handles project setup and quality control, and self-service tools for teams that want more direct control over annotation workflows.

What differentiates Appen is scale and geographic reach built up over two decades, along with experience running large, multilingual annotation pipelines for major tech companies. It has faced more competition recently from newer data-labeling and RLHF-focused firms, and has gone through layoffs and revenue declines as some big clients moved work in-house or to competitors, but it remains one of the longest-running and most recognized names in the space.