Evidence for creative work in the age of AI.
IPLYR determines whether protected works were used to build AI models and documents each finding to the standard expected by courts, insurers, and licensing counterparties.
Sources: Authors Guild and Susman Godfrey on the Anthropic settlement; ChatGPT Is Eating the World AI Copyright Case Tracker; European Commission, AI Act Article 53. IPLYR is not counsel of record in these matters.
Was this work used to train that model?
IPLYR issues a written forensic opinion on whether a specific work is present in a specific model. Each engagement returns a binary finding, EXTRACTION CONFIRMED or NO EXTRACTION DETECTED, delivered as a signed, hash-anchored evidence bundle prepared for use before the court.
Methodology draws on peer-reviewed extraction and membership-inference research. Prepared under the Daubert / FRE 702 framework. Admissibility in any particular proceeding remains the court's determination. Full technical brief available to counsel and opposing experts under MNDA.
Illustrative bundle. The finding reflects the test's output at the reporting threshold set for the engagement. Legal conclusions remain with counsel.
IPLYR is the evidence-and-attribution layer for the AI-content economy.
Choose the path that fits your work. The forensic record underneath is the same; the deliverable changes.
Per-work forensic evidence reports
Run the Litmus Test on a specific work and a specific model. Receive a binary verdict, an evidence bundle documented to the Daubert factors, hash-anchored and signed, sized for motion practice and case-portfolio underwriting.
- Binary verdict + signed evidence bundle
- Damages anchor aligned to Bartz settlement floor
- Case-portfolio underwriting briefs for funders
Training-data audits & licensing intelligence
Audit a training corpus for copyrighted exposure before regulators or plaintiffs do. Quote a deal against 270+ tracked comparators. The same forensic engine, applied upstream.
- EU AI Act Article 53 audit & summary draft
- Training Data Price Index + BATNA brief
- Continuous monitoring across catalog scale
Tamper-evident. Independently timestamped. Signed.
Each engagement produces a written forensic opinion accompanied by a cryptographic evidence bundle prepared to withstand a motion to exclude and a hostile cross-examination.
Tamper-evident
Every artifact in the report is committed to a cryptographic ledger. Later alteration of any prompt, response, or file is detectable on inspection.
Independently timestamped
The record is anchored to an external, third-party timestamp. The date the evidence existed does not depend on IPLYR's servers or clock.
Signed and attributable
Each bundle carries a signed provenance seal that discloses the signer, the signer's trust level, and the methodology version applied.
A signed seal is a verifiable record of provenance. It is not a substitute for the chain-of-custody discipline maintained by counsel. Specific cryptographic primitives and version numbers are disclosed to counsel and opposing experts under MNDA.
The engine knows what the courts just ruled.
IPLYR's fair-use scoring and jurisdiction model update as rulings land.
Bartz v. Anthropic
Training on purchased books was fair use; a library of pirated copies was not. Settled in 2025 for $1.5B, about $3,000 per work across about 482,000 works.
Kadrey v. Meta
Fair use on summary judgment where plaintiffs failed to develop a market-harm record.
GEMA v. OpenAI
Memorization is permanent infringement under EU law, not transient copying.
Holdings are summarized. Consult the full opinions. Nothing on this site is legal advice.
Bring us a work and a model. We will tell you what the evidence shows.
IPLYR is a forensic evidence platform. We do not provide legal advice or represent parties in litigation.