DATA AND ALGORITHM TRANSPARENCY
Last updated: May 17, 2026
At Sentilis we value trust over attention. This page summarizes, in plain language, what we do with data and how the systems that affect what you see and publish work. It complements — not replaces — the Privacy Notice and the Terms and Conditions.
1. A Transparent Business Model
We do not sell your data. We do not share it with data brokers, programmatic ad networks or third parties for advertising-profile building. Our business model is based exclusively on:
- The optional Pro Plan subscription (USD $12/month).
- A commission (5%–8%) on Transactions that happen in Market.
Our interests are aligned with your success: we only grow if you grow.
2. Creator Sovereignty (Zero Lock-in)
We believe your audience and your work belong to you, not to the Platform.
- Local-first storage. Your Content is stored primarily on your computer as Markdown (
.md) files. We act as the engine that compiles, publishes and distributes. - Frictionless export. If you decide to leave Sentilis, there is no complex process: the original files are already on your disk. You can also download your subscriber list as
.CSVfrom the dashboard at any time, subject to data-protection law (respecting unsubscribes and marketing consents). - Real portability. Your Content does not depend on proprietary formats. If Sentilis disappeared tomorrow, your files would still be readable by any text editor.
3. What Data We Use and Why
For each purpose we operate under the principle of minimum data necessary:
| For this… | We use this data | Legal basis |
|---|---|---|
| Create and maintain your account | Identification and contact | Contract |
| Charge and disburse payments | Transactional and tax data | Contract and tax law |
| Prevent fraud and abuse | IP, device, transactional history | Legitimate interest and law |
| Improve the product | Aggregated/anonymized usage events (Cloudflare Web Analytics) | Legitimate interest / consent |
| Newsletters and product updates | Consent |
For exhaustive detail, retention periods and sub-processors, see the Privacy Notice.
4. How the Algorithm Works
Our recommendation system is deliberately simple and explainable. It prioritizes, in this order:
- Explicit subscriptions. What you follow shows up first.
- Chronology. We value publication order. New content gets seen.
- Diversity and limited discovery. A small, declared share is reserved for content outside your subscription graph, without amplifying content by “addictive engagement” metrics.
What we do NOT do:
- We do not use behavioral-manipulation patterns (aggressive infinite scroll, notifications designed to compulsively re-engage).
- We do not prioritize content by advertising capacity.
- We do not hide content based on lawful political or artistic opinion.
5. Content Moderation
We use a hybrid (AI + human review) system to detect and handle illegal content and violations of our rules (CSAM, terrorism, fraud, hate speech, IP infringement).
- Automated decisions with material effect (suspension, content takedown, monetization restrictions) are escalated to human review before execution, except in cases of imminent harm (e.g., CSAM, active fraud).
- You have the right to appeal any decision by writing to
appeals@sentilis.me. We respond within a reasonable timeframe. - We publish a semi-annual transparency report with aggregate moderation metrics: number of takedowns, reasons, requesting countries, appeals received and reversed.
6. Artificial Intelligence
- Mandatory labeling. Creators must label content substantially generated by AI, as required by emerging regulation (including the EU AI Act).
- We do not train models on your private Content. We do not use Creator Content to train our own or third-party AI models, except with explicit, opt-in consent.
- Explainable internal AI. When we use AI for moderation or anti-fraud, we keep auditable records and human review over material cases.
7. Security and Vulnerability Reporting
We apply encryption in transit and at rest, least-privilege access control, backups and continuous monitoring. We accept responsible vulnerability reports at security@sentilis.me and publicly recognize researchers who help improve the Platform (Hall of Fame).
8. Government Requests
We only hand over data to authorities upon a valid court order or legal request addressed to the correct entity. Where law allows, we notify the affected user. Aggregate statistics on these requests will be published in the transparency report.
9. How to Hold Us Accountable
- Personal data:
privacy@sentilis.me - Moderation appeals:
appeals@sentilis.me - Security:
security@sentilis.me - Legal:
legal@sentilis.me
If you believe we are not living up to these commitments, write to us. Transparency only matters if it is enforceable.