What is this AI ethics blog about?

This blog by Rafael Darío Amador Pérez covers AI ethics, algorithmic bias auditing, explainable AI, AI transparency, and human-AI communication from a Caribbean and Latin American scholarly perspective. Posts engage the EU AI Act, UNESCO AI Ethics Recommendation, NIST AI RMF, and the Humanoide Framework for mandatory pre-deployment bias auditing.

Who is Rafael Darío Amador Pérez?

Rafael Darío Amador Pérez is a Dominican writer, researcher, and retired Lieutenant Colonel of the Ejército de República Dominicana. He is the author of Humanoide en la Torre de Babel, which proposes Neural Analysis and the Neurological Birth Certificate as binding global standards for mandatory pre-deployment AI bias auditing.

What is the Humanoide Framework for algorithmic bias auditing?

The Humanoide Framework is a three-phase mandatory pre-deployment audit methodology: Phase 1 audits training data for historical bias; Phase 2 scans the model black box using saliency maps, TCAV, and perturbation stress tests; Phase 3 simulates real-world social impact before deployment. The result is an Ethical Certification or Legal Veto.

Is this blog available in Spanish?

Yes. Rafael Darío Amador Pérez publishes in both Spanish and English. The Spanish blog at darioamadorperez.com/blog covers the same AI ethics topics for Dominican and Latin American university audiences. The book Humanoide en la Torre de Babel is available in Spanish on Amazon.

How can universities use this content for AI ethics courses?

Faculty can use these blog posts as supplementary readings for AI ethics, technology law, and algorithmic accountability courses. The book Humanoide en la Torre de Babel is available for course adoption at Hispanic-Serving Institutions. Contact the author directly for exam copies and bulk institutional pricing.

HomeRafael Darío Amador PérezBlog en Español

English Research Blog

AI Ethics, Algorithmic Bias & Human-AI Communication

Original English-language research by Rafael Darío Amador Pérez on AI governance, neural analysis, explainable AI, and the Caribbean and Latin American perspective on algorithmic accountability.

AI Regulation for Medical Devices: EU AI Act, MDR, and FDA §524B

Medical AI systems face dual compliance under the EU AI Act high-risk classification and the MDR conformity regime. The FDA §524B adds a third axis for US-deployed cyberdevices. Here is what independent auditing must look like.

Read post

Algorithmic Accountability

Deploying AI Without a Bias Audit Is Algorithmic Negligence

Organizations deploying AI systems in education, healthcare, or public services without prior independent bias auditing are not making a technical oversight — they are making a legal and ethical choice with foreseeable consequences.

Read

AI in Higher Education

AI in Higher Education: What Universities Must Do Before Deploying AI Systems

Universities are among the most aggressive adopters of AI systems for admissions, grading, and student services — and among the least prepared to audit them. Here is the framework they need.

Read

Latin America & AI

Artificial Intelligence in Latin America: The Caribbean Governance Gap

The Dominican Republic launched a national AI strategy in 2023. The governance framework needed to make it ethical has not yet been built. This post describes what it requires.

Read

AI Fairness

AI Fairness Is Not a Feature — It Is a Legal Obligation

AI fairness is routinely framed as a technical goal or a brand attribute. It is neither. It is a legal obligation that existing regulatory frameworks have failed to enforce — and that failure has documented victims.

Read

Human-AI Collaboration

Human-AI Collaboration: Why the Partnership Model Requires a Pre-Deployment Audit

Genuine human-AI collaboration requires that humans can trust the AI system they are collaborating with. That trust cannot exist without documented independent pre-deployment auditing.

Read

Neuroscience & AI

Neuroscience and Artificial Intelligence: What Neural Networks Reveal About Machine Bias

Artificial neural networks are modeled on biological ones — imperfectly. Understanding the neuroscience behind neural network architecture reveals why AI bias is structural, not incidental.

Read

AI Accountability

AI Transparency and Accountability: The Legal Framework Gap

Every major AI governance framework — EU AI Act, UNESCO Recommendation, NIST RMF — includes transparency requirements. None of them are enforceable against the conflict of interest they are designed to resolve.

Read

Explainable AI

Explainable AI: Why Transparency in Neural Networks Is a Legal, Not Just Technical, Requirement

Explainable AI (XAI) is framed as a technical aspiration. Neural Analysis argues it is a legal prerequisite — no high-impact AI system may be deployed if its decision process cannot be explained, audited, and challenged.

Read

AI Ethics

AI Ethics and Artificial Intelligence: What the Current Frameworks Miss

AI ethics as a field has produced extensive guidelines, recommendations, and principles. It has produced almost no enforceable obligations. This post examines what genuine AI ethics requires and why it has not been built.

Read

Algorithmic Bias

Why Mandatory Algorithmic Bias Auditing Is the Minimum Bar for Responsible AI

Algorithmic bias is the most underestimated structural risk in modern artificial intelligence. Mandatory pre-deployment auditing, grounded in Neural Analysis, is the only mechanism capable of ensuring AI systems do not automate discrimination at scale.

Read