AI Regulation Tracker / Regional hub

Latin America AI Regulation Tracker

The AI laws, rules and court decisions across Latin America that change what a professional must do, organized by country. We update this as new rules land.

Maintained by The Leveraged Years Regulatory Research · Last updated July 10, 2026

Brazil

Brazil Receita SC 193: AI and Lei do Bem CriteriaReceita Federal Solucao de Consulta 193, on Simples Nacional, says AI and automation work can count as innovation for Lei 11.196 incentives, on conditions.
Brazil Women's Ministry Sets Staff GenAI RulesPortaria 134/2026 binds only Ministry of Women staff and contractors in Brazil: human validation of AI output, a disclosure notice, and Copilot by name.
CADE Closes Amazon Anthropic Gun Jumping ReviewCADE says its Tribunal closed the Amazon and Anthropic review on 23 September and limited the ruling to the contracts examined. We have not read the vote.
Brazil Sends AI Companion Review to ANPD, SenaconReferral, not a finding: 13 named AI companion apps sent to the ANPD and Senacon for assessment under the ECA Digital.
Resolucion No. 001644 del 31 de julio de 2026, Ministerio de Salud y Proteccion Social (Colombia)Binding. This is an issued resolution, not a consultation or a guide. It repeals Resolucion 2654 de 2019 and sets a twelve-month transition for providers already registered to deliver teleme
Resolucion No. 89-26 del Consejo de Gabinete, approving the Estrategia Nacional de Inteligencia Artificial de Not binding on private parties. This is a Cabinet Council resolution adopting a strategy, and SENACYT describes it as containing objectives, programs, actions and priorities. It creates no c
Despacho Decisorio no 3/2026/SFI, Processo no 00261.004804/2026-54, Agencia Nacional de Protecao de DadosBinding on the addressee. This is a preventive measure issued inside an open supervisory proceeding, not a final sanction and not guidance. Discord had ten business days to appeal to the Sup
Request by the Secretaria Nacional de Direitos Digitais, Ministerio da Justica, to the Policia Federal and theNot binding, and nothing is blocked by it. This is a request from one arm of the executive to the federal police and to ANPD, each of which decides for itself whether and how to act.
Agravo de Instrumento no 5013085-69.2025.4.03.0000, TRF-3, Quarta Turma, agravo internoBinding between the parties. The Fourth Panel unanimously gave partial relief on internal appeal: the fine stands at 5 percent of the value of the principal claim, but it is now owed by the
Acordo de cooperacao tecnica TSE-AGU sobre inteligencia artificial nas Eleicoes 2026Not binding. This is a technical cooperation accord between two institutions, and the document it commissions is a guia de boas praticas, an orientation guide. It creates no new electoral of
Brazil's STF treats an AI avatar as the banned speakerAn STF order applies the electoral deepfake rule, which bars synthetic audio or video of a real person used to harm or favour a candidacy even with that person's authorisation. Consent is not a defence where the rule protects a process.
Federal court treats an AI-assisted defence as juridically non-existentThe filing was struck, default followed and a bad-faith fine of 10 percent was imposed. First instance, binding inter partes only.
Brazil bar suspends two lawyers over AI prompt injectionBrazil's Paraiba Bar suspends two lawyers for hidden prompt-injection instructions planted to hijack a court's drafting AI.
Doctors liable for AI careCFM Resolution 2.454 makes physicians own AI-assisted decisions, effective Aug 26.
AI 'smart toys' referred to regulatorsECA Digital enforcement referral over child risk.
Brazil TRT-2 Sanctions Lawyer for AI-Fabricated Case LawA Sao Paulo labor court fined a party over AI-fabricated jurisprudence and referred counsel to OAB-SP.
Labor court fines an AI citationA lawyer and company fined for AI-invented precedents, referred to the bar and prosecutors.
Election deepfake banDeepfakes banned, AI content labeled, a 72-hour synthetic-content blackout.
Human oversight of court AINo solely-automated decisions; courts must supervise AI.
ANPD live AI sandboxThree firms test AI governance under the data authority.
Facial-recognition enforcementBinding biometric rules coming; enforcement against 23 clubs.
Bar puts AI on partnersSao Paulo makes partners answerable for juniors' AI use.
The AI Act (PL 2338)Risk-tiered AI law stalled in the Chamber; where it stands.
Insurance AI sandboxSUSEP's permanent sandbox is the route for AI in claims.
AI-training copyright fightThe bill would make firms pay authors to train on their work.
Brazil PL 3066 AI child-CSAM penaltiesThe Senate approved criminal penalties for AI and deepfake child-abuse imagery, awaiting presidential sanction.
Brazil PL 5695 AI deepfake crimeA fast-tracked bill would make AI deepfakes of women a Maria da Penha crime; not yet passed.
Parana fines lawyer R$20M for AI citeA court fined a lawyer for citing a nonexistent AI-fabricated precedent and referred him to the bar.
Ibero-American data rules add AIThe shared regional standard now covers AI, neurodata, biometrics and automated-decision auditing as nonbinding reference law.
Brazil's STJ Bars AI-Generated Report as EvidenceBrazil's top court excluded an AI-made report as unreliable criminal evidence, an early high-court signal that counsel must verify AI output before filing.
Brazil opens an AI telecom consultationAnatel seeks evidence on AI in telecom through 7 September. It is a consultation, not a new compliance rule.
ANPD monitors AI intimate-content risksANPD says Decrees 12.975/2026 and 12.976/2026 are in force and is monitoring platform duties and AI-manipulated intimate-content risks. This is not a general AI Act.

Peru

Law 31814 September deadlineA September 10 AI compliance deadline for six regulated sectors.
Deepfakes as a criminal aggravatorLaw 32314 raises sentences when AI is used to defraud, defame or abuse.
Peru mandates an AI Officer for every public bodyPeru's ENIA 2026-2030 makes an AI Action Plan and an AI Officer mandatory across the public sector, in force May 2, 2026.

Chile

Chamber of Deputies advances the deepfake bill, 128-2Boletin 17795-19 was approved in general at the first constitutional stage on 22 July 2026. It is not law and imposes no duties yet.
A new data-protection regimeLey 21.719 creates an agency, breach notice and fines, from December 1, 2026.
A risk-tiered AI billAn EU-style AI systems bill cleared the Chamber and is in the Senate.

Colombia

Court voids a fabricated-citation rulingThe Supreme Court annulled a decision built on nonexistent case citations.
AI-data ground rulesThe SIC requires a privacy impact study before AI touches personal data.
Child-safety duties for accessible AI systemsDecree 769 is in force. It reaches certain child-accessible AI systems and requires scoped safe design, risk analysis and semiannual reporting.

Ecuador

A binding AI-data standardThe SPDP guarantees the right against solely automated decisions.
Ecuador COIP AI-CSAM bill (committee stage)A National Assembly committee took up a Criminal Code reform to criminalize AI-generated child sexual abuse material; a signal, not yet law.

Argentina

Buenos Aires Province gates AI purchases before they happenDecreto 742/2026 creates a mandatory AI Registry and prior clearance for AI development, procurement and renewal across the provincial administration.
Buenos Aires Province sets binding public-sector AI rulesDecree 742/2026 sets binding rules for how the Province of Buenos Aires develops, buys and runs AI.
Face data is sensitiveThe AAIP treats facial-recognition and biometric data as sensitive.

Bolivia

A sanctioning AI billA bill would give AGETIC oversight and restrict mass facial recognition.

Venezuela

An ethics code, a law aheadA voluntary AI ethics code signals a coming binding law.

Mexico

Yucatan Makes Misuse of AI Software a CrimeYucatan's Decree 214/2026 adds a new penal code offence for AI-generated harmful material, plus AI aggravators on fraud and blackmail.
Mexico New Data Law Adds AI Automated-Decision DutiesMexico's new Federal Data Protection Law gives people a right to human review of automated and AI decisions.
Mexico Bill Criminalizes AI Deepfake ImpersonationMexico's justice committee approved a Penal Code chapter that would criminalize AI deepfakes used to impersonate people.
Mexico Copyright Law: Consent to Use a Voice in AIMexico's Federal Copyright Law reform requires express written authorization, with pay, to use a person's image or voice in any AI system.

Trinidad and Tobago

Courts set binding AI filing rulesThe Trinidad and Tobago judiciary issued a practice direction setting binding rules for AI use in court filings.

Panama

Panama creates AI governance bodies by executive decreeExecutive Decree No. 36 of 7 May 2026 declares critical and emerging technologies a national priority and creates the bodies behind a National AI Strategy.

Costa Rica

Costa Rica's judiciary issues binding generative AI rulesCircular of the Secretariat of the Supreme Court No. 31-2026 sets ethical rules and guidelines for the use of generative AI across the judiciary.

Dominican Republic

Dominican competition regulator addresses algorithmic collusionPRO-COMPETENCIA issued a public recommendation report to the Senate on algorithmic collusion and pricing under the competition bill.

Cuba

Cuba Labour Code Limits Workplace AI SurveillanceCuba's Labour Code, Ley 189/2026, published 18 September 2026, says workplace AI and automated surveillance cannot breach privacy or data protection.

Regional

EU-Mercosur access testData and AI compliance become the practical test of EU market access.
Declaration of Santo Domingo AI roadmapMore than 20 governments adopted a nonbinding AI governance roadmap for 2026 to 2027.

Recent Latin America coverage

Brazil ANPD Monitors ChatGPT, Claude and GeminiANPD opened two monitoring processes on 21 August 2026 naming seven generative AI tools. The 22 notified agents must answer within ten business days.
Brazil Lets Courts Use AI to Hunt Debtor AssetsProvimento 255 lets Brazilian courts use AI for asset research and drafts in execution, but bars it from decisional acts. In force around 19 September.
Brazil ANPD Consults on Its 2027-2028 AI AgendaTomada de Subsidios 04/2026 on a draft resolution approving the ANPD regulatory agenda. Draft, unnumbered and unsigned. Four of seventeen items are AI. Contributions close 16 October 2026.
Argentina Court Denies Fees Over AI-Faked CitesA binding order in this specific case: the two lawyers lose their fees for the appeal and were referred to the bar ethics tribunal. It is not a supreme-court precedent binding on all Argentine courts, and it is an intermediate appellate decision, so read...
Brazil ANPD Note Recommends Probe of X Grok DeepfakesInternal enforcement recommendation. The note proposes opening a proceeding; a proceeding had not been decided or concluded as of the note's date
Brazil PL 6237: Executive AI Bill Attached to PL 2338Pending. Presented December 8, 2025; attached (apensado) to PL 2338/2023 on April 6, 2026; now moving jointly (Tramitando em Conjunto) under priority regime, subject to plenary review. Not enacted
Chile Boletin 17.795-19: the deepfake "integridad digital" billPending. First constitutional stage, general discussion in the Chamber of Deputies; latest recorded floor action 13 July 2026, left pending. Source: tramitacion record.
Colombia CONPES 4144 Sets National AI Governance PolicyNon-binding. A CONPES document is a public-policy roadmap that commits government entities to planned actions and budgets. It is not a law or regulation and does not by itself impose enforceable obligations on private companies.
Colombia Top Court Sets Rules for AI Use by JudgesBinding. A ruling of the Constitutional Court reviewing a tutela action. It resolves the case and sets criteria the judicial branch must observe when using AI.
Invima AI Circular 1000-018-2026 and Contract ReachSigned and issued. Ten of ten electronic signatures recorded, signed by the Director General
Colombia Judiciary Limits Judges to Approved AI ToolsBinding. It is an administrative act (acuerdo) that applies to magistrates, judges, judicial employees, and administrative staff of the Judicial Branch. It is not a statute, but within the judiciary it is mandatory internal law.
Colombia PL 043/2025S bill would name MinCiencias AI authorityDraft, en tramite. Positive first-debate ponencia filed 3 December 2025; first debate pending in the Sixth Constitutional Committee. No legal force.
Paraguay Ley 7.593/2025: Automated-Decision Rights in DP LawEnacted and published 27 Nov 2025; enters into force 27 Nov 2027 (Article 57).
Uruguay Decreto 276/025: AI Testing & Data Spaces FrameworkIn force; committee has 120 days from publication to issue its rules and minimum-content documents.