In recent days, Colombia has seen several updates in terms of AI regulation. The most relevant updates on the matter are the following:

1. Law 2502/2025: The Colombian Congress enacted Law 2502, amending Article 296 of the Criminal Code to introduce a specific aggravating circumstance for cases of identity fraud committed through the use of artificial intelligence (AI), including the creation of deepfakes and digital identity impersonation. Under this new provision, fines for such offenses may be increased by up to one third when AI tools are used. The law also provides definitions for key concepts such as deepfake, identity, and image, and strengthens the legal protection of personal rights against technological manipulation.
   
   In addition, Law 2502 sets out guidelines for the development of public policies aimed at preventing and controlling the misuse of AI in identity fraud. These include the establishment of an ethical framework, intersectoral collaboration, training in cybersecurity and digital ethics, the development of detection technologies, transparency in AI algorithms, international cooperation, and the implementation of rapid response protocols. This amendment will take effect one year after the law’s sanction and promulgation.
    
2. Draft Bill 043-25: This draft bill filed by the Colombian Government intends to regulate the AI in Colombia in a comprehensive manner. 

Overview

The Draft Law No. 043 of 2025 aims to establish a comprehensive regulatory framework for the development, use, and governance of Artificial Intelligence (AI) in Colombia. Its primary objective is to ensure the ethical, responsible, competitive, and innovative deployment of AI, while safeguarding fundamental rights and promoting sustainable development.

The law would apply to all individuals and entities, public or private, involved in any stage of the AI system lifecycle—design, development, deployment, operation, commercialization, or use—when the system is developed, used, or has effects in Colombia, or uses data from Colombian sources.

Key Principles

The regulation is guided by principles such as human oversight, diversity and inclusion, social and environmental well-being, ethics, transparency, responsible innovation, privacy, proportionality, respect for fundamental rights, environmental protection, economic development, technological sovereignty, adaptability, multi-stakeholder collaboration, and fair competition.

Risk-Based Classification

AI systems are classified into four risk categories:

• Critical Risk: Systems that may substantially violate fundamental rights or public interest (e.g., subliminal manipulation, social scoring, real-time biometric identification by authorities). Their use is generally prohibited except under strict conditions.
• High Risk: Systems with significant potential impact on health, safety, or rights (e.g., in education, employment, public services, law enforcement, justice). These require conformity assessments, risk management, transparency, and human oversight.
• Transparency Obligations: Systems that interact with people or generate/manipulate realistic content (e.g., deepfakes) must clearly inform users of their artificial nature.
• Minimal or No Risk: Systems outside the above categories, encouraged to adopt voluntary best practices.

Governance and Supervision

The Ministry of Science, Technology, and Innovation is designated as the national authority for AI, responsible for technical guidance, coordination, and oversight. A multi-sectoral committee and a National Advisory Council of AI Experts will support governance, policy updates, and inter-institutional collaboration.

Innovation and Education

The law promotes research, innovation, and the creation of regulatory sandboxes for safe experimentation. It mandates the integration of AI education at all levels, with a focus on inclusion, regional development, and workforce transition.

Responsibilities and Sanctions

Clear responsibilities are established for developers, providers, implementers, and users of AI. Administrative and, in the future, criminal sanctions are foreseen for non-compliance, with a preference for corrective and educational measures before punitive actions.