Skip to main content
NOPE
In Effect Regulation AI Safety

Peru AI Regulations

Supreme Decree No. 115-2025-PCM - Regulations of Law No. 31814 (AI Promotion Law)

Peru's first comprehensive AI regulatory framework, inspired by EU AI Act. Establishes three-tier risk-based approach: prohibited uses, high-risk systems (including healthcare), and low-risk/acceptable AI. First general AI regulation in Latin America. Requires human oversight, transparency, and risk assessments for high-risk AI including healthcare applications.

Jurisdiction

Peru

Enacted

Sep 9, 2025

Effective

Jan 22, 2026

Enforcement

Presidency of the Council of Ministers (PCM) through Secretariat of Government and Digital Transformation (SGTD); National Centre for Digital and AI Innovation (CNIDIA)

Published September 9, 2025; effective January 22, 2026 (90 business days after publication); phased implementation 1-4 years by sector

El Peruano Official Gazette

Why It Matters

First comprehensive AI law in Latin America. Healthcare AI classified as high-risk requiring human oversight, transparency, and risk documentation. Establishes precedent for risk-based AI regulation in region. Relevant for any AI health service operating in or targeting Peru. Human-in-the-loop requirements ensure AI cannot autonomously make healthcare decisions.

Recent Developments

First comprehensive AI regulatory framework in Latin America. Published September 9, 2025; effective January 22, 2026. Phased implementation over 1-4 years depending on sector and entity size. National AI Sandbox established for controlled regulatory testing.

At a Glance

Applies to

Healthcare AI

Who Must Comply

  • Public sector entities (central, regional, local governments)
  • Private sector developers and deployers of AI systems
  • Organizations using high-risk AI for healthcare, employment, social programs, credit, critical infrastructure
  • All AI systems used in Peru (with exceptions for narrow commercial applications)

Obligations fall on:

Safety Provisions

  • Prohibited: manipulative practices, unauthorized mass surveillance, predictive policing, autonomous lethal systems
  • High-risk AI (healthcare, employment, social programs, credit) requires human oversight
  • Mandatory transparency disclosures about AI purpose and capabilities
  • Risk assessments documented before high-risk AI deployment
  • Human oversight mechanisms - staff must be able to stop, correct, or invalidate AI decisions
  • Register maintained with system principles, data sources, algorithmic logic, ethical impacts
  • Ethics guidelines required (to be issued within 180 days)
  • National AI Sandbox for regulatory testing

Compliance & Enforcement

Key Dates

Sep 10, 2025

Four Final Complementary Provisions take effect

Jan 22, 2026

Main regulation takes effect (90 business days after publication)

Jan 22, 2027

Central government 1-year compliance deadline

Jan 22, 2028

Regional government compliance (2-3 years)

Jan 22, 2029

Local government compliance (3 years)

Penalties

Penalties pending regulatory determination

View on map

Peru

Focus Areas

Algorithmic accountability
Active safeguards required

Cite This

APA

Peru. (2025). Supreme Decree No. 115-2025-PCM - Regulations of Law No. 31814 (AI Promotion Law).

Related Regulations

In Effect PE

Peru AI Law

First AI-specific law in Latin America. Privacy protection throughout AI lifecycle. Accountability for fundamental rights violations.

In Effect SV

El Salvador AI Law

First comprehensive AI law in Latin America. Promotes AI development while establishing ethical principles and governance framework. Creates the National Agency for Artificial Intelligence (ANIA) to oversee AI development and regulation.

In Effect CA-ON

Ontario Bill 194

Ontario's first AI-specific legislation regulating public sector use of AI systems. Requires accountability frameworks, risk management, disclosure, and human oversight. Also addresses cybersecurity and digital information affecting minors under 18.

In Effect MX

MX Safe Digital Environments Decree

Mexican presidential decree, binding nationwide, that prohibits digital platforms from giving children and adolescents access to, exposure to or recommendations of seven listed content categories, including suicide, self-harm and eating disorders, and requires age assurance beyond self-declaration, parental supervision tools, strictest-by-default settings for new accounts, periodic risk assessments and a 24/7 law-enforcement contact point.

Failed CA

C-63

Would have established Digital Safety Commission with platform duties for seven harmful content categories including content inducing children to harm themselves. Required 24-hour CSAM takedown.

In Effect BR

Brazil ECA Digital

Comprehensive child digital safety law applying to any IT product or service directed at or likely to be accessed by minors in Brazil, with extraterritorial reach.

All regulations See what's likely relevant to your deployment

Last updated January 22, 2026. Verify against primary sources before relying on this information.