LANDR Expands Stream Fraud Detection System

LANDR has announced the expansion of its stream fraud detection system, developed in partnership with Mila. Mila is a non-profit artificial intelligence research institute based in Montreal and founded by Yoshua Bengio. LANDR, which says its platform serves more than 8 million musicians, reported that the volume of artificial streams flagged by Spotify across music it distributed fell by 95% between Q1 2025 and Q2 2026. The company attributes this decrease not only to the Mila system, but to the implementation of multiple anti-fraud initiatives.

The detection system includes models that assess accounts through profile and behavioral data, alongside listening patterns inconsistent with a genuine audience. The system was expanded to identify additional risk signals earlier in the distribution process. These signals consist of unusually high release frequencies, catalogs appearing overnight, and account clusters sharing device or payment details. In addition, LANDR is introducing identity verification for accounts that show frequent release activity.

Machine learning functions as one layer of LANDR's anti-fraud system, which also incorporates audio recognition technologies from ACRCloud and Ircam Amplify. Detected activity is reviewed by an anti-fraud team prior to enforcement. LANDR stated that human review sits between detection and enforcement because an unusual pattern in data is not proof of fraud.