Europe Data privacy

EU Interim Privacy Rules for OTT Scanning

For OTT chat, webmail, and in-app messaging providers serving EU users, Regulation (EU) 2026/1881 changes the compliance boundary between communications confidentiality, voluntary content detection, and data retention. Adopted in July 2026, it allows eligible providers to detect, report, and remove online child sexual abuse material under narrowly defined safeguards. The interim regime applies until 3 April 2028 and requires providers to document processing, false-positive rates, retention, safeguards, and cross-border data transfers.

Published:11/05/2026 Updated:11/05/2026

1. Regulatory focus

Regulation (EU) 2026/1881 creates a temporary derogation from the confidentiality rules in the ePrivacy Directive for certain number-independent interpersonal communications services. Providers may voluntarily detect, report, and remove online child sexual abuse material and detect child solicitation, but audio scanning is excluded and communications using, or scheduled to use, end-to-end encryption are outside scope. Processing must remain necessary, proportionate, and compliant with the GDPR.

2. Business impact

Content scanning can no longer be treated as an isolated trust-and-safety feature. Providers need an auditable legal basis, documented detection logic, human review and redress processes, and deletion controls. Within six months of entry into force and every 31 January thereafter, providers must report processing volumes, GDPR legal grounds, international transfers, identified cases, complaints and remedies, false-positive rates, retention periods, and safeguards to the competent supervisory authority and the Commission.

3. Operating recommendations

Create a communications-processing inventory segmented by product, EU market, encryption status, and data type, identifying which chats, attachments, or traffic data may enter a detection workflow. Track false-positive rates, correction times, escalation evidence, and retention separately for hash matching, text classifiers, and human review. Before enabling detection for EU users, prepare the reporting schema, transfer assessment, user complaint channel, and permanent deletion workflow for data that is no longer strictly necessary.

Frequently Asked Questions

Must a provider scan end-to-end encrypted chats when using hash matching?
No. The regulation excludes communications to which end-to-end encryption has been, is being, or will be applied, and it does not require providers to decrypt communications. Product and privacy documentation should distinguish readable content, end-to-end encrypted content, and metadata processing so that the provider does not describe all messaging flows as subject to the same detection controls.
How long may detection data be retained?
Data may be retained only for as long as strictly necessary for detection, reporting, or removal. Where no suspected abuse is identified, the data should be permanently deleted once it is no longer necessary. Even where a suspected case is identified, the regulation sets a maximum of 12 months from detection, subject to separate EU or national preservation obligations.
Who should prepare the annual compliance report?
The report should be owned cross-functionally rather than by trust and safety alone. Product, privacy, legal, security, and data engineering teams should maintain processing volumes, legal grounds, transfers, case counts, false positives, complaint outcomes, and retention controls. Assign a regulatory submitter and an internal reviewer, with every reported figure traceable to logs, tickets, and deletion records.
This article is for informational purposes only and does not constitute legal advice.

Related products

Need compliance guidance?

Contact us for guidance on target markets, message scenarios, and sending routes.

Get in Touch
Privacy Policy