DOJ–RealPage Proposed Consent Decree: Federal Antitrust Remedy Reaches an Algorithm's Training Data, Not Just Its Live Inputs
The Antitrust Division's proposed consent decree in United States v. RealPage (with landlord defendant Pinnacle Property Management Services), noticed at 91 FR 59304 on September 18, 2026, bars Pinnacle from using any revenue-management product that ingests external non-public competitor data at runtime, pools or discloses it — "or that was trained on it." As CyberEyeQ Weekly #39 (September 24, 2026) reads it: "The reach into the training corpus is the part to read twice: it is the clearest federal statement yet that an algorithm's provenance, not just its live inputs, can be the remedy's target."
Why this matters beyond landlord pricing software: the remedy establishes that a downstream deployer's liability exposure attaches to how its vendor's model was built — the provenance of training data — not merely to what the deployed system does in production. For enterprises deploying third-party revenue-management, pricing, or other algorithmic decisioning tools, the consent-decree template implies diligence duties up the vendor chain: what competitor data went into the model, when, and under what lawful basis. Tunney Act comments close on or about November 17, 2026, so the final judgment text may still move.
This federal development lands alongside the state track tracked in this topic's earlier cycles — California AB 325's algorithmic-pricing disclosure regime and the multi-state consumer-protection pincers around algorithmic pricing (see DOJ–RealPage Proposed Consent Decree: Federal Antitrust Remedy Reaches an Algorithm's Training Data, Not Just Its Live Inputs for the state-law layer). The combined effect: algorithmic pricing is now regulated from both ends — input provenance (DOJ) and output disclosure (states).