LANDR expands fraud detection built with Mila, reports 95% drop in flagged streams

Montreal-based distribution and music creation platform LANDR announced on September 22 an expansion of the streaming-fraud detection system it built with Mila, the Quebec non-profit AI institute founded by Yoshua Bengio.

The company says the volume of artificial streams flagged by Spotify across the music it distributes fell 95% between Q1 2025 and Q2 2026 — a figure LANDR attributes to several fraud-prevention measures rather than to the Mila models alone. One model assesses accounts using profile and behavioural data; another looks for listening patterns inconsistent with a genuine audience.

Additional risk signals now run earlier in the distribution process: unusually high release frequency, catalogues that appear overnight, clusters of accounts sharing devices or payout details, and metadata imitating established artists. Flagged activity goes to human review before any action, the company says.

The stakes are commercial: since April 2024 Spotify charges labels and distributors a per-track fee for flagrant artificial streaming. LANDR says its work follows the priorities of IFPI's Streaming Integrity Initiative, published on September 14.