AI Regulation Tracker / Legislative signal

Ecuador's Justice Committee takes up a bill to criminalize AI-generated child sexual abuse material

On July 8, 2026, the National Assembly's Justice Committee formally opened study of a Criminal Code reform that would create a distinct offense for child sexual abuse material produced with artificial intelligence. This is a committee-stage proposal, not yet law.

By Anthony Guerriero · The Leveraged Years Regulatory Research · Published July 13, 2026 · Last updated July 13, 2026

The short version

Ecuador's Comisión de Justicia y Estructura del Estado, a committee of the National Assembly, formally took up a bill on July 8, 2026 that would amend the Código Orgánico Integral Penal to criminalize child sexual abuse material generated with artificial intelligence and other information technologies. This is the earliest legislative stage. The committee has opened study of the bill; it has not been debated on the floor, passed, or signed, and it imposes no obligation today. We are covering it as a signal, not a rule. It matters because it is another jurisdiction moving to close the gap between existing child-protection law, which assumes a real victim image, and synthetic material a generative model can produce. For platforms and AI image tools that reach Ecuador, it is an early marker of where liability is heading.

Regulatory Briefing

Instrument: Proyecto de Ley Orgánica Reformatoria al Código Orgánico Integral Penal (COIP) para la Tipificación de Conductas Vinculadas a la Explotación Sexual de Niñas, Niños y Adolescentes mediante Inteligencia Artificial y otras Tecnologías de la Información
Authority: Asamblea Nacional del Ecuador, Comisión de Justicia y Estructura del Estado
Jurisdiction: Ecuador
Bindingness: Non-binding at this stage (SIGNAL); a committee-stage proposal, not enacted law
What Changed: The committee formally took up ("avocó conocimiento") the bill for study on July 8, 2026.
Who This Affects: If enacted: prosecutors, courts, platforms, and generative-image tools operating in Ecuador. Today: a directional signal only.
Current Status: Under committee study; not debated, passed, or signed
Sponsor: Assemblywoman Carolina Dueñas (per the Assembly's published notice)
Primary Source: Asamblea Nacional, asambleanacional.gob.ec
Last Verified: July 13, 2026

Most criminal codes written before generative AI assume that child sexual abuse material depicts a real child. That assumption is now a loophole. A model can produce synthetic imagery that no existing photograph underlies, and prosecutors in several countries have found that their statutes do not clearly reach it. Ecuador's Justice Committee is the latest to try to close that gap.

What happened, precisely

On July 8, 2026, the Comisión de Justicia y Estructura del Estado announced that it had formally taken up two bills for study. One is a reform of the Código Orgánico Integral Penal, Ecuador's criminal code, that would define and criminalize conduct tied to the sexual exploitation of children through artificial intelligence and other information technologies. The Assembly's own notice records the step as "avocó conocimiento," the procedural act of a committee opening its study of a bill.

That is the whole event, and the limits matter. Taking up a bill is not passing it. The committee has not reported the bill out, the full Assembly has not debated or voted on it, and the President has not signed anything. There is no new duty on anyone in Ecuador today, and there is certainly no cross-border obligation. Anyone describing this as a law Ecuador has passed is wrong about the stage.

Why it belongs on the tracker as a signal

We hold this to the signal standard rather than the action standard, and it clears that bar. It is a formal, primary-sourced legislative step by a national body on a duty that generative AI is actively reshaping. For a compliance team at a platform or an AI image company, the value is directional. Ecuador is joining a widening set of jurisdictions, including several in Latin America, that are writing synthetic-media child-protection offenses into their criminal law. The specifics will change as the bill moves, but the direction of travel is consistent, and it is the direction a global content-moderation and model-safety program should already be built for.

Questions professionals are asking

Has Ecuador criminalized AI-generated child sexual abuse material?

Not yet. As of July 8, 2026, a National Assembly committee has only opened study of a bill that would do so. It has not been passed by the Assembly or signed into law, and it creates no obligation at this stage.

Should a platform or AI-image company change anything today because of this?

Nothing is legally required by this step. The practical read is directional: more jurisdictions are writing synthetic-CSAM offenses into criminal law, so a moderation and model-safety program serving Latin America should already be built to that standard.

RELATED BRIEFINGS

Browse the full AI Regulation News tracker

Informational analysis for working professionals, not legal advice. Confirm how any rule applies to your situation with qualified counsel in the relevant jurisdiction.