AI Ethics & Responsible Use Policy
The principles guiding how MAP uses automation and AI tools in its verification services.
Last updated: October 2026
1. Purpose
MAP Verified™ context records are produced by automated, rule-based extraction code; no generative AI model writes record text. This policy explains the principles we follow so that our use of automation and AI tools stays transparent, accountable, and useful. See the AI Disclosure Policy for specifics.
2. Guiding principles
Transparency — every MAP context record comes with a breakdown of the specific observable evidence behind it, not a numeric rating. Honesty about limits — where our data doesn’t support a confident finding on a particular dimension, we say so in the record rather than filling the gap. No autonomous high-stakes decisions — MAP context records are informational; we do not present them as certified fact, legal findings, or financial advice. Security — we take reasonable measures to protect our extraction code and systems against unauthorized access and modification.
3. How context records are produced
MAP Verified™ context records are compiled from data our extraction code can actually read from the article or publisher in question — headline characteristics, outbound references, domain attribution, and content-specificity indicators. Full detail on the methodology is available on our methodology page.
4. Prohibited uses
We do not use AI or automated systems for surveillance, generating disinformation, producing deepfakes, or making discriminatory decisions about individuals.
5. Human oversight
MAP’s extraction code reads observable information from eligible public content, such as headline, metadata, and outbound links. The application assembles context records from those extracted signals and writes the resulting record to the MAP ledger automatically. Individual automated context records do not receive human editorial review before publication. MAP is run by its founder, who maintains the extraction code and methodology, investigates reported errors, and makes publisher-certification decisions separately; no publisher has been certified yet. No AI model determines truth, credibility, political bias, publisher certification, or consequential decisions. The founder uses AI coding and writing assistants to build and maintain the website and its code. We do not currently claim independent third-party certification of our AI governance program (such as ISO/IEC 42001); this policy reflects our internal operating principles.
6. Reporting concerns
If you believe a MAP context record is inaccurate, incomplete, or was produced improperly, contact us at legal@mediaalertpress.com and we will review it.
7. Policy review
We review this policy periodically as our systems and tools evolve. Last reviewed: October 2026.
8. Contact
Media Alert Press LLC, 15050 Elderberry Lane, Fort Myers, FL 33907. Email: legal@mediaalertpress.com