LANDR Reports 95% Reduction in Artificial Streams Flagged by Spotify as It Expands AI-Powered Fraud Detection

23 September 2026

Technology developed with Mila, the Quebec AI Institute, now enables LANDR to flag suspicious activity earlier in the release process, advancing the same priorities outlined in IFPI’s new Streaming Integrity Initiative, backed by the three major label groups and leading independent music organizations.

MONTREAL, September 23, 2026—8AM ET—LANDR today announced the expansion of its AI-powered streaming fraud detection system, originally developed in collaboration with Mila – Quebec AI Institute, a leading AI research institute based in Montreal, founded by Turing Award winner and deep learning pioneer Yoshua Bengio. The enhanced system flags more suspicious activity before releases reach streaming services, giving LANDR's anti-fraud team an earlier opportunity to review it.

The expansion comes as the music industry increases its focus on streaming fraud and the role distributors can play in preventing it. LANDR’s is advancing many of the same priorities outlined in IFPI’s new Streaming Integrity Initiative, which calls for stronger safeguards across five areas: Verify, Vet, Act, Share, and Measure.

LANDR's developers spent six months working alongside Mila researchers to develop two complementary machine-learning models: one that assesses accounts using profile and behavioral data, and another that examines listening activity for patterns inconsistent with a genuine audience.

Between Q1 2025 and Q2 2026, the company saw a 95% reduction in artificial streams flagged by Spotify, following the rollout of multiple fraud-prevention initiatives. The machine-learning system developed with Mila, deployed in February 2026, is one part of that broader strategy.

LANDR has since expanded its capabilities to identify additional risk signals earlier in the distribution process, including unusually high release frequency, catalogs that appear overnight, clusters of accounts sharing devices or payout details, and metadata designed to impersonate established artists.

LANDR’s anti-fraud team reviews flagged activity before taking action. Accounts confirmed to be fraudulent can be closed and blocked from returning, while confirmed cases help LANDR continue improving its detection models.

Finding fraud by hand at our scale was a needle-in-a-haystack problem. The first generation of the model turned that into something we could actually manage,” said Tommy Desilets, CTO, LANDR. “Since then, we’ve continued improving the system as we identify new patterns of fraudulent activity. Now we’re moving those capabilities earlier in the distribution process, so our team has better signals before suspicious releases reach streaming services.

Tommy Desilets, CTO, LANDR

Fake streams are plays generated without genuine listener intent, often through bots, stream farms, or compromised accounts. When undetected, they can divert revenue away from artists and rightsholders with legitimate audiences.

An artificial stream isn’t just somebody gaming a platform. It competes for money from the same royalty pool that pays musicians who are trying to make a living, said Pascal Pilon, co-founder and CEO, LANDR.

Pascal Pilon, co-founder and CEO, LANDR

But fighting fraud also means protecting legitimate artists from being wrongly flagged. An unusual pattern in the data isn’t proof of fraud. A catalog imported in one batch or a track that suddenly finds an audience may trigger additional review while being entirely legitimate. That’s why human review sits between detection and enforcement.

We want AI helping our people make better decisions at scale, not replacing judgment,” Pilon said. "Sometimes a song suddenly taking off is exactly that: an artist having a really good day. We’re rooting for them, not trying to put a damper on their success.

Pascal Pilon, co-founder and CEO, LANDR

Machine learning is one layer of LANDR’s broader fraud-prevention system. LANDR also uses audio-recognition technologies to identify potentially problematic recordings before distribution. ACRCloud fingerprints recordings against commercial catalogs, while technology from IRCAM checks for altered recordings, including sped-up, pitch-shifted and re-recorded copies, as well as acoustic markers associated with synthetic generation.

LANDR is also rolling out identity verification for accounts with frequent release activity, allowing its team to apply additional checks on higher-risk activity without creating unnecessary friction for most independent artists.

Artists using LANDR declare AI-generated content when submitting a release, with that information included in the metadata delivered to streaming platforms.

For artists choosing a distributor, fraud protection is increasingly part of the value of the service itself. LANDR’s distribution offering combines these safeguards with a flat annual plan, in-house support, and a free catalog importer that allows artists to move their music to LANDR from other distributors while retaining existing play counts and playlist placements.

Our goal is simple: stop fraud earlier without creating more barriers for legitimate artists, Pilon said.

Pascal Pilon, co-founder and CEO, LANDR

About LANDR
LANDR is an AI-powered music platform that helps artists create, distribute, and monetize their work. Its tools and services span distribution, promotion, monetization, mastering, samples, plugins, and collaboration, serving a global community of more than 8 million musicians building sustainable careers on their own terms. LANDR also owns both Reason Studios and Synchro Arts. Learn more at landr.com.

Media Contact
Karine Fortin
Head of External Communications, LANDR
kfortin@landr.com
+1-514-699-4952