CONITEC, IETS, and CENETEC are Latin America's three most influential HTA agencies. None has a public position today on artificial intelligence in pharmaceutical evidence. Here is what that means for market access teams in the region.

In April 2024, the ISPOR HTA Roundtable conducted an informal survey of 18 members of HTA agencies from North America, Latin America, Australasia, and Europe to measure internal use of generative AI in those organizations. The result was revealing: at the time of that analysis, only NICE had issued a public position on the use of AI in evidence generation [1].

None of the Latin American agencies had, or has today, a clear framework on how they will evaluate evidence built with artificial intelligence. And yet, the pharmaceutical industry is already starting to use generative AI tools in parts of its dossier processes [2].

The ground where credibility will be decided

That gap between what the industry is doing and what agencies have defined is not an administrative detail. It is the ground on which the credibility of pharmaceutical evidence in LATAM will be decided over the next two to three years.

1. CONITEC, IETS, and CENETEC: profiles and current state on AI

CONITEC (Brazil): the most demanding agency in the region

CONITEC is Latin America's most solid HTA methodological reference. Created in 2011 under Federal Law 12.401, it has more than 100 multidisciplinary professionals and evaluates technologies for the SUS, the public health system covering more than 200 million Brazilians [3]. Its methodological guidelines for economic evaluations are detailed, with explicit criteria on systematic reviews, cost-effectiveness models, and budget impact.

CONITEC has no public position today on AI in evidence generation. Brazil approved Bill 2.338/2023 on general AI regulation in the Senate in December 2024, now under review by the Chamber of Deputies [4]. That framework establishes principles of transparency and risk classification inspired by the European AI Act, but does not specifically address HTA processes or pharmaceutical evidence.

IETS (Colombia): rising methodological rigor

The Institute of Health Technology Assessment plays a central role in Colombian health system coverage, reference prices, and clinical practice guidelines. Its methods are based on ISPOR and EUnetHTA standards, and its technical team actively participates in international HTA networks.

Like CONITEC, IETS has no public position on AI in evidence. Colombia introduced Bill 043/2025 on AI regulation in July 2025, with urgent government status in September 2025 [5]. The bill follows a risk-based approach similar to the European AI Act and incorporates OECD principles, but without direct application to HTA processes.

CENETEC (Mexico): an expanding role

The National Center for Health Technology Excellence was created in 2004 with an initial focus on medical devices, and has progressively expanded its role toward drug evaluation for Mexican public health system formularies [6]. It operates as a technical advisor to the Ministry of Health and the General Health Council on technology coverage and prioritization.

Mexico also has no specific CENETEC position on AI in HTA. Between 2021 and 2024, legislative initiatives were introduced to regulate AI in the country, but none has been approved, and CENETEC has published only non-binding digital health guidelines [7].

2. What the HTAi Global Policy Forum 2025 said about the future

In 2025, Health Technology Assessment international (HTAi) gathered representatives of HTA agencies, pharmaceutical industry, payers, and patients at its Global Policy Forum to discuss the use of AI in HTA activities. The conclusions are directly relevant to LATAM [2].

The forum confirmed that the life sciences industry is already actively exploring generative AI tools for clinical evidence generation and is starting to use them in the HTA submission process, including the construction and adaptation of dossiers.

The forum's position on how HTA agencies should advance was clear:

AI implementation in HTA activities must be gradual, starting with small, less complex, and repetitive tasks before expanding to higher-risk applications.
Agencies must create sandbox-type environments to learn, adjust, and build trust in AI tools proactively.
The forum's consensus was unanimous: the time to act is now. Waiting for total certainty before developing frameworks means falling behind an industry that is already advancing.

For CONITEC, IETS, and CENETEC, that message has direct implications: agencies that do not define their position on AI in evidence soon will find themselves evaluating dossiers built with tools their evaluators are not trained to identify or question.

3. The only global reference available today: what NICE established

In August 2024, NICE published its Position Statement on the use of AI in evidence generation (ECD11) [8]. It was the first such document issued by a top-tier global HTA agency, and it quickly became a reference for the industry and other agencies.

The five central principles of the NICE framework are:

Mandatory transparency: document which AI tools were used, how, and under what assumptions.
Required validation: AI methods must be contrasted with alternative methods when possible.
Non-negotiable human oversight: AI augments expert judgment, it does not replace it.
Justified use: AI should only be used when it adds clear value compared to established methods.
Recognized high risk for comparative effect estimation: requires sensitivity analysis and triangulation with available clinical evidence.

Canada's Drug Agency followed NICE guidelines with its own Position Statement in 2025. EMA has an AI workplan for 2023-2028. FDA published its draft guidance for AI in regulatory decisions in 2025. The global pattern is unequivocal: reference agencies are building frameworks. LATAM has not yet.

4. Why the silence is not neutral: the risk to the industry

The absence of formal positions at CONITEC, IETS, and CENETEC does not protect those who use AI without rigor. Evaluators at these agencies are professionals trained in HEOR methodology, with access to the same international standards as their European peers. A dossier with AI-generated evidence without process documentation, without validation, and without transparency exposes the entire evidence strategy to methodological integrity concerns.

The Putnam Associates report (2025) documents this precisely: although several pharmaceutical companies already have internal initiatives to explore AI in HEOR, they rarely use it in real dossiers for fear of non-acceptance by agencies [9]. The use of AI in evidence remains uncommon and often undeclared.

Two opposite but equally problematic consequences

Teams that use AI without declaring it expose themselves to integrity concerns if the process comes to light during evaluation. Teams that avoid AI out of fear compete at a growing disadvantage in time and cost against those integrating it with rigor.

The way out is not to wait for agencies to publish their positions. It is to prepare today with the standards that already exist.

5. What LATAM market access teams must do now

Adopt NICE standards as the minimum floor

NICE ECD11 guidelines are the most complete standard available for AI use in HTA evidence. Building processes that meet their criteria, transparency, validation, human oversight, justified use, is the best possible preparation for when CONITEC, IETS, and CENETEC define their own positions, which will follow the same model.

Document the process with as much rigor as the result

The most important change AI introduces is not the final output, it is process traceability. A dossier documenting how AI was used, with which tools, what validations were performed, and how expert oversight was maintained, can be defended before any evaluator today and when the rules arrive.

Train teams before it becomes mandatory

The HTAi Global Policy Forum was explicit: the time to act is now. Organizations waiting for rules to be published before training teams will arrive late. CONITEC, IETS, and CENETEC evaluators already participate in ISPOR and know the international standards. Market access teams must know them too.

The vacuum will close. The question is whether your team will be ready.

CONITEC, IETS, and CENETEC will face growing pressure to define their position on AI. Evaluation demand is rising, resources are limited, and AI tools to accelerate processes are increasingly accessible. When positions arrive, they will follow the model NICE already established: AI permitted, under strict conditions.

Pharmaceutical companies building that methodological capability today, not just using AI, but using it well, with rigor and traceability, will be the ones arriving prepared at that moment.

How we help

At Quantus, we support market access and HEOR teams to build evidence processes that are rigorous and aligned with the international standards that already exist. If you want to know how to prepare your team for what is coming in LATAM, write to us.

References

[1] Fleurence R, et al. Generative Artificial Intelligence for Health Technology Assessment: Opportunities, Challenges, and Policy Considerations. Value in Health. 2025;28(2):175-183. doi:10.1016/j.jval.2024.10.3846

[2] HTAi Global Policy Forum. Advancing the use of artificial intelligence in health technology assessment activities. Int J Technol Assess Health Care. 2026. doi:10.1017/S0266462324001983

[3] CONITEC/INAHTA. National Committee for Technology Incorporation. Available at: inahta.org/members/conitec

[4] Library of Congress. Brazil: Senate Advances Discussions on Bill to Regulate AI Use. May 2025. Available at: loc.gov/global-legal-monitor

[5] AmCham Colombia. Proyecto de ley sobre inteligencia artificial: elementos clave de la regulación en Colombia. August 2025. Available at: amchamcolombia.co

[6] Caetano R, et al. A Perspective on the use of Health Technology Assessment in Brazil and Mexico. Value in Health. 2015. doi:10.1016/j.jval.2015.09.2936

[7] Chambers and Partners. Digital Healthcare 2025, Mexico: Trends and Developments. Available at: practiceguides.chambers.com

[8] NICE. Use of AI in evidence generation: NICE position statement (ECD11). August 2024. Available at: nice.org.uk/corporate/ecd11

[9] Putnam Associates. Acceptance of Artificial Intelligence in Evidence and Dossier Developments by HTA bodies: Challenges and Opportunities. 2025. Available at: putassoc.com