PL
AI&Scale

AI Transparency Policy

Last updated: 5 June 2026

AI&Scale uses AI to support research, drafting and product experiences — with mandatory human oversight for published claims and with clear disclosure where AI materially shapes output. This page explains what we do and what we do not do.

Where we use AI

Drafting assistance (outlines, rewrites, translation support) under editorial prompts.

Summaries and structural transformations of our own content.

Product experiences such as writing assistance (“Ghostwriter”) and content recommendations.

Where we do not use AI as a decision-maker

We do not use AI to make binding decisions about individuals (employment, credit, eligibility) on the platform.

We do not present AI-generated output as professional legal, tax, financial, investment or HR advice.

Human editorial control

Editorial responsibility remains human. Publishing requires human review for framing, sourcing discipline and factual plausibility.

When AI-assisted content is published, we disclose it in a clear, reader-facing way.

EU AI Act transparency

For interactive AI features, we inform users they are interacting with an AI system.

For AI-assisted or synthetic media, we apply clear labels in line with applicable transparency obligations.

Labels you may see

AI-assisted

AI supported drafting or transformation; a human editor reviewed and approved the final publication.

Synthetic image

An image generated or materially modified using AI tools.

Ghostwriter

An interactive AI writing assistant intended to help structure thinking and drafts, not to replace professional judgment.

Commitments

  • →We prioritize trust over speed: we prefer fewer, higher-confidence claims over unverified volume.
  • →We distinguish analysis, opinion and frameworks; we mark material updates and corrections.
  • →We continuously refine disclosures as regulations, product features and risk posture evolve.

User-provided data & AI

  • →We do not intentionally ask users to submit sensitive personal data into AI tools.
  • →If AI features process user-provided text, we aim to minimize retention and limit use to service provision.
  • →We do not use user-provided content to train external AI models unless explicitly agreed.