Centered ballot with blue crayon

Freedom of expression in the 2026 elections

Under what conditions was freedom of expression exercised during Costa Rica's 2026 electoral process? To answer that question, we monitored 7,301 publications from 45 national and regional media outlets between April 1, 2025 and February 1, 2026. This note summarizes the main findings of the report.

publicaciones monitoreadas
7.301
medios nacionales y regionales
45
meses, de abril de 2025 a febrero de 2026
10
recomendaciones hacia 2028
82

Freedom of expression encompasses the right to seek, receive and disseminate information. During an electoral process, the exercise of this right is essential because it allows citizens to learn about the candidates, compare their proposals and make informed choices. That is why the report examines the environment in which the campaign developed, access to information, political party financing, misinformation, attacks on journalists and media outlets, and the use of new technologies.

The monitoring was conducted using a digital platform created for this purpose, which used artificial intelligence to continuously track media websites based on keywords. The report "Freedom of Expression in Costa Rica: 2026 General Elections" measures the availability of published information and its impact on the exercise of freedom of expression, not its journalistic quality.

We detail the main findings of the study.

Information focused on who was winning, not on government proposals

Between September 1, 2025 and January 16, 2026, 1,461 news stories were published and only 31 referred in their headline to campaign proposals. In contrast, voter intention polls concentrated 206 news stories and became the main electoral narrative.

The presentation of candidacies was defined by newsworthiness criteria. Those that did not generate interest due to their proximity to power, their capacity to generate conflict, or their position in polls were not covered. In fact, eight presidential candidacies had no discussion of their plans in the monitored media. This phenomenon was visible in issues of high citizen concern, such as security, which was the contextual matter with the greatest presence (281 of 1,329 news stories), but no systematic and comparative analysis of the candidacies' proposals on that issue was recorded. In the last week of the campaign, a study cited by the press indicated that 47% of undecided voters were still seeking more information about the candidacies (Sandoval, J. D., 2026).

ExplorerLas veinte candidaturas en tres medidas

Compare a las veinte personas candidatas por menciones propias, por artículos sobre sus propuestas o por gasto facturado. Si no tiene JavaScript en su dispositivo, las cifras están en las páginas 29, 30 y 48 del reporte.

Loading the explorer…

Added to this was the fact that the most visible figure on the country's news agenda was not participating in the race.Then-president Rodrigo Chaves and the Executive Branch were present in 1,922 publications (26.3%), above the TSE (22.1%) and the Legislative Branch (20.5%). The most mentioned candidacy was Laura Fernández's and appeared in only 528 news stories.

The recorded news stories focused on conflicts between the branches of government, which limited discussions of government plans from different political parties.

Legislative Assembly: an election overshadowed by the presidential race

Of the 2,696 news stories analyzed in the information access chapter, only 282 referred to the election of the 57 deputations.

The distribution was highly unequal: 45% for PPSO, 17.7% for PLN and 9.2% for Frente Amplio, while eight parties, four of them provincial, received no mention in nine months. Less than one in ten of those pieces offered substantive information about proposals, with the provinces most affected being Limón, Puntarenas and Guanacaste.

The pattern repeated itself at the close of voting. Of the 95 news stories about results that were recorded, only four addressed the legislative result independently of the presidential result and the quota and remainder system was hardly explained.

Excessive dependence on the TSE

The Supreme Electoral Tribunal played a fundamental role as a source of information and guarantor of electoral organization. However, the report identifies excessive dependence on this institution to communicate and explain the process. For example, information about ballots and the placement of parties was provided at only one moment and parties did not replicate it in their campaigns. The clarification that the digital ID was not valid for voting was disseminated solely by the Tribunal. In addition, communication directed to the 130,000 people who were incorporated into the registry was scarce, and it was recorded that 24,675 people were left out by failing to request their ID card in time (Arrieta, E., 2025). Voting abroad showed the same pattern. 91 Polling Stations were deployed in 49 consulates in 42 countries, but coverage of the process was minimal.

This concentration also revealed that when the TSE became simultaneously organizer, referee and party to public controversies, its communications had to also respond to the defense of its own legitimacy.

Judicialization of controversies and delayed adoption of decisions

Political differences resulted in judicialization: 831 news stories reported formal cases and electoral injunctions predominated (108), political involvement (88) and immunity and removal processes (49).

The procedures demonstrated institutional capacity to intervene; however, in several cases final decisions were not available before voting. For example, in August 2025 there were 135 pending charges for political involvement, some filed since 2019, and the precautionary measure on religious propaganda, issued on January 16, 2026, was adopted when there were fifteen days left before the election without a final resolution of the case before February 1.

The ability to investigate conduct is not sufficient if the information needed to understand its consequences arrives after citizens have made their decision.

Disinformation and artificial intelligence: transformation faster than regulation

The bills presented by the TSE in 2023 to update rules on propaganda, social media, and artificial intelligence were not approved before September 30, 2025, so the Court faced the process without updated regulatory tools.

80% of the population expressed concern about the effects of disinformation on the electoral process (Córdoba, D., 2025). However, only 150 of the 7,301 articles directly addressed the phenomenon. The information analyzed shows that disinformation spread where official information was scarce or ambiguous, particularly around the digital ID, the composition of voting receiving boards, vote counting, and party financing.

The 2026 elections also marked a turning point in the use of artificial intelligence. Nine of the 20 parties with presidential candidates acknowledged to the TSE having used these tools. Tools that expanded access to information, such as the Informed Voter project assistants, became available as of January 2026, too close to the vote for wide dissemination.

Insufficient transparency in financing

The existence of a regulatory framework does not, by itself, guarantee the availability of adequate and timely information. Only one article with partial spending figures was recorded, positioning campaign investment as the most opaque area in terms of information access. Without that information, it is not possible to determine whether a candidacy's visibility was due to discursive merits or investment capacity.

Attacks against journalists

Of the 263 articles about attacks against journalists and media, the vast majority were not linked to the electoral process and only three cases of verbal aggression occurred within the election campaign. Their frequency, however, should not be the only risk indicator. A study published in 2025 found that 54.8% of participants stated that practicing journalism in the country is dangerous (Boza, G., Jiménez Alvarado, O., Rodríguez López, J., & Ramírez Varas, S., 2025). The low number of cases reflects the absence of published aggressions, not the absence of aggressions.

What worked

The electoral calendar was met and coverage was continuous, with participation from national and regional media. The TSE demonstrated technical capacity and institutional willingness to protect rights. The roster consultation tool facilitated the voting process, official presentations ensured a space for all 20 candidacies; and a December 2025 decision mandated guaranteeing access to electoral information for persons deprived of liberty. The media, for their part, demonstrated that the capacity to analyze government plans exists when editorial decisions allow for it. The process concluded with recognition of the results by the defeated candidacies.

Recommendations for 2028

The main challenge for freedom of expression in future electoral processes is not solely ensuring that information exists, but ensuring that it contributes to informed participation.

The report formulates 82 recommendations, ordered according to when they should be implemented: 36 before the call, 39 during the campaign, and 7 after election day. Of these, 26 are directed to the TSE, 20 to political parties, 20 to media outlets, 12 to civil society, and 4 to the Legislative Assembly.

TSE 26

Antes de la convocatoria 17

  1. II-TSE-1Atender el rezago en la resolución de denuncias por beligerancia política
  2. III-TSE-2Diseñar una campaña dirigida a quienes votan por primera vez
  3. III-TSE-3Diseñar una campaña dirigida a personas mayores y con discapacidad
  4. III-TSE-4Ampliar la información sobre la conformación de las JRV
  5. III-TSE-5Informar sobre el voto en el extranjero antes de la convocatoria
  6. III-TSE-8Alinear la modernización de los documentos de identidad con el calendario electoral
  7. III-TSE-9Acordar con los medios un porcentaje mínimo de espacio para las propuestas de cada candidatura
  8. IV-TSE-3Habilitar un portal de seguimiento del financiamiento partidario
  9. IV-TSE-4Dotar a las oficinas de fiscalización del presupuesto y del personal necesarios
  10. V-TSE-1Ampliar la difusión de las herramientas educativas para identificar noticias falsas
  11. V-TSE-2Crear un mecanismo de monitoreo y alerta temprana
  12. VI-TSE-1Establecer un protocolo de atención a periodistas
  13. VI-TSE-2Incluir en la regulación de propaganda pautas explícitas sobre el trato a periodistas y medios
  14. VI-TSE-3Establecer, con universidades, ONG y el COLPER, un mecanismo de registro de incidentes
  15. VII-TSE-1Mantener los chatbots de información electoral y publicarlos al menos dos meses antes de la votación
  16. VII-TSE-3Definir cuándo la actividad de un creador de contenido constituye propaganda electoral
  17. VII-TSE-4Establecer la trazabilidad y el etiquetado obligatorio de la propaganda generada o asistida por IA

Durante la campaña 7

  1. II-TSE-2Publicar la resolución de los amparos electorales en el momento en que se adopta
  2. II-TSE-3Ampliar la difusión de las actividades del IFED con personas menores de edad
  3. III-TSE-1Mantener la herramienta interactiva de localización de la JRV
  4. III-TSE-6Destacar la participación en el extranjero durante la jornada electoral
  5. IV-TSE-1Publicar de forma proactiva, accesible y periódica el desembolso de fondos a los partidos
  6. IV-TSE-2Solicitar periódicamente a los partidos el detalle de la inversión de campaña
  7. VII-TSE-5Formalizar y reportar públicamente los acuerdos con las plataformas de redes sociales

Después de la jornada 2

  1. III-TSE-7Incluir los resultados del voto en el extranjero desde el primer boletín
  2. VII-TSE-2Evaluar el impacto real del uso de IA en la información electoral
Partidos políticos 20

Antes de la convocatoria 5

  1. IV-PP-2Declarar por anticipado los posibles conflictos de interés
  2. V-PP-2Mantener cuentas oficiales verificadas e informar cuáles son las auténticas
  3. VI-PP-1Adoptar códigos de conducta que prohíban las agresiones a periodistas
  4. VI-PP-2Capacitar a voceros y equipos de comunicación en el manejo de las tensiones
  5. VII-PP-1Adoptar códigos de conducta sobre el uso ético de la IA en la campaña electoral

Durante la campaña 14

  1. II-PP-1Asumir un rol activo en la difusión de información sobre la celebración de las elecciones
  2. II-PP-2No excluir a los medios de comunicación de los actos de campaña
  3. III-PP-1Difundir las propuestas de gobierno y los perfiles de las candidaturas durante la campaña
  4. III-PP-2Fortalecer la comunicación provincial para la elección legislativa
  5. III-PP-3Replicar la información sobre las papeletas y la ubicación de los partidos
  6. III-PP-4Evitar la judicialización de casos que obedecen a la dinámica política de la contienda
  7. IV-PP-1Publicar de forma voluntaria, periódica y desagregada la inversión publicitaria
  8. IV-PP-3Evitar las denuncias públicas y la judicialización de casos sin elementos sustanciados
  9. V-PP-1Verificar la información antes de difundirla y no amplificar contenidos no confirmados
  10. V-PP-3Publicar de forma proactiva la información sobre el financiamiento y el uso de los recursos de campaña
  11. VI-PP-3Condenar pública y oportunamente las agresiones cometidas por seguidores
  12. VII-PP-2Señalar de forma visible los contenidos generados o asistidos por IA
  13. VII-PP-3Transparentar la inversión y la segmentación de la pauta digital
  14. VII-PP-4Aclarar la relación entre el partido y los creadores de contenido que lo apoyan

Después de la jornada 1

  1. V-PP-4Reportar el cumplimiento de los pactos firmados en materia de desinformación y discursos de odio
Medios de comunicación 20

Antes de la convocatoria 4

  1. III-MC-7Desarrollar con el TSE planes comunicacionales sobre el calendario electoral
  2. IV-MC-1Desarrollar coberturas explicativas del modelo de financiamiento en su conjunto
  3. V-MC-1Publicar más sobre las herramientas y cursos que enseñan a detectar noticias falsas
  4. VII-MC-1Continuar desarrollando chatbots y herramientas de análisis

Durante la campaña 14

  1. II-MC-1Publicar notas comparativas entre las propuestas de gobierno
  2. II-MC-3Revisar los criterios de selección de noticias para reducir la asimetría entre candidaturas
  3. III-MC-1Equilibrar la cobertura de las encuestas y la de las propuestas de gobierno
  4. III-MC-2Publicar comparaciones temáticas entre planes de gobierno
  5. III-MC-3Ampliar la información diferenciada por provincia
  6. III-MC-4Evitar que la conflictividad política domine la agenda comunicacional
  7. III-MC-5Dar servicio informativo a las personas costarricenses en el extranjero
  8. IV-MC-2Verificar con rigor las denuncias sobre financiamiento antes de publicarlas
  9. V-MC-2Verificar los datos que aportan las personas candidatas durante la campaña
  10. V-MC-3Cubrir la desinformación como tema en sí mismo y no solo mediante desmentidos puntuales
  11. VI-MC-1Sistematizar y reportar de forma constante los ataques recibidos
  12. VI-MC-2Diferenciar con claridad la violencia de la tensión política del momento
  13. VI-MC-3Colaborar con el TSE y la sociedad civil en el mecanismo de registro de incidentes
  14. VII-MC-2Etiquetar de forma visible los contenidos y análisis producidos con IA

Después de la jornada 2

  1. II-MC-2Dar seguimiento a los procesos hasta su resolución
  2. III-MC-6Reportar los hitos finales del proceso
Sociedad civil 12

Antes de la convocatoria 6

  1. III-SC-1Facilitar la comunicación con grupos específicos
  2. III-SC-4Distribuir las herramientas tecnológicas de comparación con antelación suficiente
  3. V-SC-2Adoptar el Acuerdo Nacional y el Pacto Ético Interpartidario Sitio Mata 2026 como marco de comportamiento
  4. VI-SC-2Impulsar campañas de respeto a periodistas y medios de comunicación
  5. VII-SC-1Continuar desarrollando herramientas basadas en IA sobre el contenido de las propuestas electorales
  6. VII-SC-3Ampliar los espacios de alfabetización digital

Durante la campaña 4

  1. III-SC-2Producir análisis comparativos de los planes de gobierno
  2. III-SC-3Difundir, con medios y partidos, las propuestas para la Asamblea Legislativa
  3. V-SC-1Publicar los reportes periódicos comprometidos sobre el cumplimiento de los acuerdos
  4. VII-SC-2Mantener los estudios de escucha social reforzando la transparencia metodológica

Después de la jornada 2

  1. VI-SC-1Mantener el monitoreo permanente de las agresiones a periodistas y medios de comunicación
  2. VII-SC-4Articular con universidades estudios sobre la incidencia de las redes sociales y de los creadores de contenido
Asamblea Legislativa 4

Antes de la convocatoria 4

  1. IV-AL-1Tramitar las reformas pendientes desde 2023
  2. IV-AL-2Reformar el régimen de fiscalización y de acceso al financiamiento público
  3. V-AL-1Actualizar el marco regulatorio electoral en materia de desinformación
  4. VII-AL-1Adoptar un marco regulatorio para el control del gasto y de la campaña en espacios digitales y con IA

Durante la campaña 0

None at this stage

Después de la jornada 0

None at this stage

Fuente: Garrido (2026), pp. 91-95. El código remite al capítulo del reporte donde se justifica cada recomendación.

Municipal elections will be held on February 6, 2028, and groups wishing to register must submit their application no later than February 5, 2027. This is, therefore, the opportune moment for each actor to review the recommendations that apply to them.

Declaration of interests

The author participated as a national observer accredited by the Supreme Court of Elections in the mission organized by the University for Peace. The report is an independent work and does not represent the position of any institution.

References

Arrieta, E. (December 1, 2025). 24,000 new voters failed to obtain ID in time to vote. La República. https://www.larepublica.net/noticia/24-mil-nuevos-electores-no-sacaron-cedula-a-tiempo-para-votar

Boza, G., Jiménez Alvarado, O. M., Rodríguez López, J., & Ramírez Varas, S. (2025). 3rd Survey on freedom of expression and trust in mass media in Costa Rica. Proledi-Cicom.

Córdoba, D. (November 26, 2025). UCR Report: 80% of Costa Ricans worry about disinformation and its influence on upcoming elections. CRHoy. https://crhoy.com/nacionales/informe-ucr-80-de-ticos-se-preocupa-por-la-desinformacion-y-su-influencia-en-proximas-elecciones/

Garrido V., M. (2026). Freedom of Expression in Costa Rica: General Elections of 2026. San José, Costa Rica.

Sandoval, J. D. (January 28, 2026). Almost half of undecided voters still seek more information about candidates. CRC 89.1. https://crc891.com/nacionales/casi-la-mitad-de-indecisos-aun-buscan-mas-informacion-sobre-candidatos/

Share

Related notes

AI has rights. The journalist does not

Recently, Anthropic updated its usage policies to prohibit cruel treatment toward Claude, its artificial intelligence model. The system can even terminate a conversation when the user becomes…

Read more

No room for error

Artificial intelligence can accelerate learning for those starting out or take away the work where learning happens. Costa Rica needs to know which of the two things is happening in its labor force.

Read more

One hundred meters south of the gap

AILA identifies which AI capabilities the country lacks to take advantage of artificial intelligence; the Human Development Atlas shows where. Combined, they stop being diagnostics and become a tool…

Read more