Status: DRAFT — FOR LEGAL REVIEW Version: 0.2 Date: 2026-10-18 Owner: Head of Trust & Safety Jurisdiction: Canada (operating company); EU users served under the DSA and GDPR; market neutral Applicable law: Criminal Code of Canada and Personal Information Protection and Electronic Documents Act (PIPEDA) for the operating company; for EU users, Regulation (EU) 2022/2065 (Digital Services Act), in particular Art. 15 (general transparency reporting obligations), Art. 24 (additional transparency obligations for online platforms), Art. 42 (VLOP-specific — not currently applicable but structure accommodates future growth) Reviewed by counsel: [pending] Publication cadence: Quarterly Published at: auroramembers.com/transparency
This is a template for Aurora's quarterly transparency report under DSA Art. 15 and Art. 24. Placeholder fields use the
{{placeholder}}convention and are populated at publication time by the admin transparency page (src/app/(admin)/admin/transparency/page.tsx) from themoderation_eventsappend-only log and related data sources. The template is a draft for legal review: counsel should confirm completeness against Art. 15(1) and Art. 24(1) and identify any jurisdiction-specific additions (for example, guidance from the Digital Services Coordinator of the Member State where Aurora's EU legal representative is established).
Aurora Transparency Report — {{report_period_start}} to {{report_period_end}}
Reporting entity: [Registered company name], [Province], Canada (business number [Business number]) EU legal representative: Aurora's EU legal representative under DSA Art. 13: [to be appointed] Competent Digital Services Coordinator: the DSC of the Member State where Aurora's EU legal representative is established Reporting period: {{report_period_start}} – {{report_period_end}} (calendar quarter {{quarter}}) Publication date: {{publication_date}} Prior report: {{prior_report_link}} Average monthly active recipients in the EU: {{mau_eu}} (DSA Art. 24(2) figure, calculated per Commission guidance)
1. Executive Summary
Narrative prepared by the Head of Trust & Safety. 2–4 paragraphs summarising the quarter, the biggest moderation themes, material policy changes, and headline numbers.
{{executive_summary}}
2. Orders From Authorities (Art. 15(1)(a) DSA for EU Member State authorities)
This section covers orders received from Canadian authorities and from EU Member State judicial or administrative authorities requiring Aurora to act against items of illegal content — for example, court orders to remove specific URLs or to preserve data for criminal investigations.
2.1 Orders to act against illegal content
| Jurisdiction (Canada / EU Member State / other) | Category of illegal content | Orders received | Response (median, hours) | Action taken |
|---|---|---|---|---|
| {{ms_1}} | {{category_1}} | {{count_1}} | {{median_hours_1}} | {{action_1}} |
| {{ms_2}} | {{category_2}} | {{count_2}} | {{median_hours_2}} | {{action_2}} |
| ... | ... | ... | ... | ... |
| Total | {{total_illegal_orders}} | {{overall_median_hours}} |
2.2 Orders to provide information
| Jurisdiction (Canada / EU Member State / other) | Legal basis | Orders received | Median response time | Information provided / refused |
|---|---|---|---|---|
| {{ms_a}} | {{basis_a}} | {{count_a}} | {{time_a}} | {{result_a}} |
| Total | {{total_info_orders}} |
Aurora notifies affected users of orders unless legally prohibited (see Law Enforcement Guidelines §8). {{notification_count}} affected users were notified this quarter; {{nondisclosure_count}} orders were subject to non-disclosure requirements.
3. Notices Submitted Under Art. 16 (Art. 15(1)(b) DSA)
This section covers notices submitted by users and trusted flaggers alleging illegal content on Aurora.
3.1 Notices by source
| Source | Notices received | Notices acted on | Notices rejected | Pending at end of period |
|---|---|---|---|---|
| Registered users | {{notices_users}} | {{acted_users}} | {{rejected_users}} | {{pending_users}} |
| Trusted flaggers (Art. 22) | {{notices_tf}} | {{acted_tf}} | {{rejected_tf}} | {{pending_tf}} |
| Unregistered public (web form, email) | {{notices_public}} | {{acted_public}} | {{rejected_public}} | {{pending_public}} |
| Total | {{notices_total}} | {{acted_total}} | {{rejected_total}} | {{pending_total}} |
3.2 Notices by category of alleged illegality
| Category | Notices received | Median time to decision | Action taken (most common) |
|---|---|---|---|
| Child sexual abuse material (CSAM) | {{csam_notices}} | {{csam_median}} | {{csam_action}} |
| Human trafficking indicators | {{traf_notices}} | {{traf_median}} | {{traf_action}} |
| Non-consensual intimate imagery (NCII) | {{ncii_notices}} | {{ncii_median}} | {{ncii_action}} |
| Fraud / impersonation | {{fraud_notices}} | {{fraud_median}} | {{fraud_action}} |
| Harassment / threats | {{harass_notices}} | {{harass_median}} | {{harass_action}} |
| Copyright | {{copyright_notices}} | {{copyright_median}} | {{copyright_action}} |
| Privacy violation / doxxing | {{priv_notices}} | {{priv_median}} | {{priv_action}} |
| Hate speech / incitement | {{hate_notices}} | {{hate_median}} | {{hate_action}} |
| Terrorism content (Reg. 2021/784) | {{tco_notices}} | {{tco_median}} | {{tco_action}} |
| Drugs / weapons / illegal goods | {{goods_notices}} | {{goods_median}} | {{goods_action}} |
| Illegal services | {{ill_serv_notices}} | {{ill_serv_median}} | {{ill_serv_action}} |
| Spam / platform integrity | {{spam_notices}} | {{spam_median}} | {{spam_action}} |
| Other | {{other_notices}} | {{other_median}} | {{other_action}} |
| Total | {{category_total}} | {{category_overall_median}} |
3.3 Median response time — all categories
- Acknowledgement of notice: {{ack_median}} (target: < 24 hours)
- Decision on notice: {{decision_median}} (target: < 7 days standard, < 24 hours for CSAM / trafficking / imminent harm)
3.4 Actions taken on valid notices
| Action | Count |
|---|---|
| Content removed | {{removed}} |
| Content restricted / geo-blocked / demoted | {{restricted}} |
| Content labelled | {{labelled}} |
| Content left in place (notice rejected) | {{left_in_place}} |
| Account suspended | {{account_suspended}} |
| Account permanently terminated | {{account_terminated}} |
| Referral to law enforcement | {{le_referral}} |
4. Content Moderation on Aurora's Own Initiative (Art. 15(1)(c) DSA)
This section covers moderation actions Aurora took of its own initiative — that is, not in response to a notice from a user or an order from an authority.
4.1 Actions by category
| Category | Total actions | Automated-only detection | Automated + human | Human-only |
|---|---|---|---|---|
| CSAM (Arachnid Shield hash match) | {{csam_own}} | {{csam_auto}} | {{csam_mixed}} | {{csam_human}} |
| NSFW / out-of-policy imagery | {{nsfw_own}} | {{nsfw_auto}} | {{nsfw_mixed}} | {{nsfw_human}} |
| Trafficking indicators | {{traf_own}} | {{traf_auto}} | {{traf_mixed}} | {{traf_human}} |
| Impersonation / shared photos | {{imp_own}} | {{imp_auto}} | {{imp_mixed}} | {{imp_human}} |
| Spam / bots | {{spam_own}} | {{spam_auto}} | {{spam_mixed}} | {{spam_human}} |
| Fraud / scams | {{fraud_own}} | {{fraud_auto}} | {{fraud_mixed}} | {{fraud_human}} |
| Terms-of-service violations (other) | {{tos_own}} | {{tos_auto}} | {{tos_mixed}} | {{tos_human}} |
| Total | {{own_total}} | {{auto_total}} | {{mixed_total}} | {{human_total}} |
4.2 Types of restriction imposed
| Restriction | Count |
|---|---|
| Content removed | {{own_removed}} |
| Content restricted (demoted / hidden from search) | {{own_restricted}} |
| Monetization / feature restriction | {{own_monetization}} |
| Warning issued | {{own_warning}} |
| Account suspended | {{own_suspended}} |
| Account permanently terminated | {{own_terminated}} |
| Referral to law enforcement | {{own_le}} |
5. Complaints Under Art. 20 (Internal Complaint-Handling System)
| Metric | Value |
|---|---|
| Complaints received | {{complaints_received}} |
| Complaints concerning content removal | {{complaints_content}} |
| Complaints concerning account suspension / termination | {{complaints_account}} |
| Complaints concerning rejected notices | {{complaints_rejected_notices}} |
| Complaints resolved in favour of complainant (decision reversed) | {{complaints_upheld}} |
| Complaints resolved against complainant (decision upheld) | {{complaints_denied}} |
| Median time from complaint submission to decision | {{complaint_median}} (target: ≤ 14 days) |
| Pending at end of period | {{complaints_pending}} |
Complaints are handled by staff not involved in the original decision, and are not decided solely on the basis of automated means (Art. 20(6)).
6. Out-of-Court Dispute Settlement (Art. 21)
| Metric | Value |
|---|---|
| Disputes referred to certified ODS bodies | {{ods_referred}} |
| Bodies involved | {{ods_bodies}} |
| Disputes resolved in user's favour | {{ods_user_won}} |
| Disputes resolved in Aurora's favour | {{ods_aurora_won}} |
| Average duration of proceedings | {{ods_avg_duration}} |
| Aurora's compliance with decisions | {{ods_compliance}} |
7. Suspensions for Abuse of Platform and Notice Systems (Art. 23)
| Metric | Value |
|---|---|
| Users suspended for frequently providing manifestly illegal content | {{art23_users_content}} |
| Persons/entities suspended for frequently submitting manifestly unfounded notices | {{art23_notifiers}} |
| Persons/entities suspended for frequently submitting manifestly unfounded complaints | {{art23_complainants}} |
| Suspensions reversed on appeal | {{art23_reversed}} |
8. Human Resources Dedicated to Content Moderation (Art. 15(1)(b), Art. 42(2))
| Metric | Value |
|---|---|
| Total moderation staff (FTE) | {{moderation_fte}} |
| In-house (Aurora employees) | {{inhouse_fte}} |
| Outsourced / contracted | {{outsourced_fte}} |
| Native language coverage | {{languages}} |
| Specialised roles (CSAM-trained, trafficking-trained) | {{specialist_fte}} |
| Training hours delivered in period | {{training_hours}} |
| Wellbeing programme coverage | {{wellbeing_coverage}} |
9. Accuracy of Automated Systems (Art. 15(1)(e))
Aurora discloses, for each automated system used for content moderation, qualitative information about purpose and indicators of accuracy where measurable.
| System | Purpose | Indicators | Accuracy metric (period) | Possible error rates |
|---|---|---|---|---|
| Arachnid Shield hash match | Detection of known CSAM | Hash match on upload | {{photodna_accuracy}} | False positive: {{photodna_fp}}; reviewed by human before permanent action (except automatic upload block) |
| Photo auto-publish | Release of a photo after a clear hash match and no classifier hold | Rule (src/lib/media/photo-route.ts) | n/a | Publishes only, restricts nobody; logged as media.auto_published; count: {{auto_published_photos}}; reversible on report |
| Perceptual image hashing | Cross-account duplicate detection (impersonation / trafficking signal) | Photo similarity across unrelated accounts | {{phash_accuracy}} | FP: {{phash_fp}}; FN: {{phash_fn}} |
| NSFW classifier | Categorisation of media | Category prediction | {{nsfw_accuracy}} | FP: {{nsfw_fp}}; FN: {{nsfw_fn}} |
| Text classifiers (minor-indicator, trafficking language, spam) | Text flagging for human review | Keyword + embeddings | {{text_accuracy}} | Reviewed 100% by human before action |
| Sanctions / PEP screening | AML compliance | Name match | {{sanctions_accuracy}} | FP: {{sanctions_fp}}; all reviewed by Compliance Officer |
| Member trust score | Informational signal shown to providers | Composite score | Not a moderation decision — no action taken on score alone | — |
Safeguards: automated-only decisions are used only where (a) they are legally required (e.g., CSAM hash match triggering upload block pending human review) or (b) they have no consequential effect on the user (e.g., search ranking). All other actions are confirmed by a human reviewer.
10. Restrictions Imposed (Art. 15(1)(b))
| Restriction type | Count |
|---|---|
| Account suspensions (temporary) | {{suspensions_total}} |
| Account permanent terminations | {{terminations_total}} |
| Content removals | {{content_removals_total}} |
| Content demotions / search-hiding | {{demotions_total}} |
| Feature restrictions | {{feature_restrictions_total}} |
| Referrals to law enforcement | {{le_referrals_total}} |
| Referrals to NCMEC / Canadian Centre for Child Protection (Cybertip.ca) | {{csam_referrals}} |
11. Preservation and Emergency Disclosures
| Metric | Value |
|---|---|
| Preservation requests received | {{preservation_received}} |
| Preservation requests honoured | {{preservation_honoured}} |
| Emergency disclosures made | {{emergency_disclosures}} |
| Average time from emergency request to response | {{emergency_response_time}} |
12. Methodology and Changes From Previous Report
Narrative describing any methodological changes, definitional clarifications, or new data sources used in this report vs. the prior one. Where a change materially affects comparability, Aurora republishes the affected prior figures on the same basis for transparency.
{{methodology_notes}}
13. Data Sources and Verification
- All figures in Sections 2–11 are drawn from the
moderation_eventsappend-only log, thereportstable, the law-enforcement request log maintained by Aurora Legal, and HR records (for Section 8). - Aggregates are computed by a reproducible ETL pipeline documented in
docs/transparency/etl.md. - Figures are reviewed by the Head of Trust & Safety and the Compliance Officer before publication.
- The report is published at auroramembers.com/transparency and archived for at least 5 years.
14. Contact
Questions or corrections: transparency@auroramembers.com (stub) Data protection: dpo@auroramembers.com Legal / law enforcement: law-enforcement@auroramembers.com
DRAFT — FOR LEGAL REVIEW. This template must be reviewed and approved by qualified legal counsel, and cross-checked against final DSA implementing acts, guidance from the competent Digital Services Coordinator, and Commission templates for transparency reporting before first publication. All {{placeholder}} fields are filled from the live data pipeline at publication time.