ARG-001

Buenos Aires Boti GenAI Chatbot (WhatsApp Citizen Services with Azure OpenAI)

Download PDF
Argentina Latin America & Caribbean Upper middle income Full Production Deployment Confirmed

Secretariat of Innovation and Digital Transformation, Government of the City of Buenos Aires (GCBA); Ministry of Human Development and Habitat (for social services integration)

At a Glance

What it does LLMs for content creation, transformation and modality conversion — User communication and interaction
Who runs it Secretariat of Innovation and Digital Transformation, Government of the City of Buenos Aires (GCBA); Ministry of Human Development and Habitat (for social services integration)
Programme Buenos Aires City Citizen Services Platform (Boti WhatsApp Chatbot)
Confidence Confirmed
Deployment Status Full Production Deployment
Key Risks Market, sovereignty and industry structure risks
Key Outcomes 2 million queries per month handled without human intervention; 82% automated resolution rate; 50% reduction in operational burden for municipal teams; 90% average monthly satisfaction rate; over 11 million vaccination appointments facilitated during COVID-19 campaign; near-100% monthly service availability; selected to train 3 Iberoamerican cities in chatbot implementation; Best Corporate Chatbot 2019 (Planeta Chatbot Awards); recognised by Facebook as international COVID-19 vaccination solution example.
Source Quality 6 sources — Government website / press release, News article / media, Report (multilateral / development partner)

The Government of the City of Buenos Aires (Gobierno de la Ciudad Autónoma de Buenos Aires, GCBA), through its Secretariat of Innovation and Digital Transformation, operates Boti, a WhatsApp-based chatbot that serves as the primary digital interface between the municipal government and its residents. Launched in February 2019, Buenos Aires became the among the first governments to use WhatsApp as an official citizen communication channel. Boti was originally built on the Botmaker platform with natural language processing capabilities and has since been upgraded with Microsoft Azure OpenAI generative AI services, integrating GPT-4o and a Retrieval Augmented Generation (RAG) architecture to enable natural-language, personalised interactions at scale.

Boti handles approximately 2 million queries per month without human intervention, covering more than 450 topics through over 70 integrations with government systems and 15 human service queues staffed by more than 250 operators for cases requiring human escalation. By 2022, the chatbot had exceeded 26 million cumulative conversations, and by the time of its generative AI upgrade in mid-2024 it had accumulated over 153 million total interactions since launch. Botmaker, the Argentine technology company that built the original platform, reported that 82 percent of all queries are resolved by the chatbot without human intervention, with only 18 percent referred to human agents.

The generative AI upgrade, announced in August 2024 and developed in partnership with Microsoft and the consulting firm Pi Data Strategy, incorporated GPT-4o through Azure OpenAI Services. The system uses a RAG pattern — described by Microsoft News Centre as the 'patrón de Generación Aumentada por Recuperación' — as the foundational element of its grounding strategy, ensuring that responses are anchored in verified government data rather than uncontrolled model outputs. The RAG architecture retrieves information from a centralised government content repository and combines it with GPT-4o's language generation capabilities to produce contextualised, conversational responses. The system includes content filtering and Azure OpenAI monitoring for quality assurance. The initial GenAI deployment focused on tourism as a lower-risk domain to test personalisation capabilities without compromising sensitive data, with the chatbot generating personalised itineraries and recommendations based on official City Tourism Board (Ente de Turismo de la Ciudad) data. The system operates in both Spanish (using the Argentine 'voseo' dialect from the River Plate region, with inclusive language and emoji use) and English to serve international visitors.

Beyond tourism, Boti provides a wide range of city services spanning public health, social protection, administrative procedures, and urban management. During the COVID-19 pandemic, Boti became a critical public health infrastructure tool, providing symptom consultation, monitoring mild cases, scheduling vaccination appointments (facilitating over 11 million vaccination appointments during the campaign), delivering test results, and issuing health certificates. The system's peak demand during the pandemic significantly exceeded WhatsApp's platform constraint of 200 messages per second, reaching over 400 messages per second for COVID-19 test result delivery, which required the deployment of overflow channels on Facebook Messenger and web platforms. Monthly conversation volumes peaked at 11 million in January 2022, compared to a pre-pandemic baseline of approximately 200,000 monthly conversations.

In the social protection domain, Boti was integrated in April 2022 as a channel for social assistance services directed at vulnerable populations, developed collaboratively by the Secretariat of Innovation and Digital Transformation, the Subsecretariat of Citizen Services, and the General Directorate of Community Outreach under the Ministry of Human Development and Habitat (Ministerio de Desarrollo Humano y Hábitat). Through this integration, citizens can request comprehensive assistance for individuals experiencing homelessness, obtain information about ministerial social programmes and Social Inclusion Centres (Centros de Inclusión Social), submit housing subsidy receipts (Subsidy 690), and inquire about payment dates for specific benefit programmes. The chatbot also provides information on disability rights and inclusion, childhood and adolescence protection, family assistance programmes, gender violence support services, addiction care, and elderly care. For social service queries, users can escalate to live operators to resolve questions related to social programmes. Previously, these consultations were only available at in-person Social Care offices (formerly Zonal offices) under the Secretariat of Social Inclusion and Immediate Care.

Additional services include driver's licence renewal and issuance, national identity document appointments, public transport schedules, real-time parking availability, vehicle violation reports with image recognition, recycling guidance, abandoned vehicle removal requests, pothole and infrastructure repair reporting, and access to emergency helplines (lines 144 for gender violence, 102 for children and adolescents, and 108 for social emergencies). Boti also features innovative engagement content including interactive children's stories ('Boticuentos'), tongue twisters, jokes, and digital sticker downloads.

The GCBA reports that Boti's generative AI integration has reduced the operational burden on municipal teams by approximately 50 percent and improved citizen satisfaction, with a 90 percent average monthly satisfaction rate. The system maintains near-100 percent monthly service availability. The development team comprises UX designers, software developers, data science specialists, journalists and communicators, and quality assurance testers, operating under Scrum and Kanban agile methodologies. The chatbot employs a service bus architecture enabling interoperability and traceability across government systems, with georeferencing capabilities using APIs such as parking rule maps for location-specific data.

Key government officials overseeing the initiative include Diego Fernández, Secretary of Innovation and Digital Transformation; Pedro Alessandri, Undersecretary of Smart City; Julieta Rappan, General Director of Digital Channels; and Macarena Blasi, Chief of Staff. Fernando López Iervasi, President of Microsoft for Hispanic South America, has been quoted stating that generative AI 'brings us the opportunity to resolve needs immediately using language learned since childhood.' Due to its success, the City of Buenos Aires was selected to provide professional support to other Iberoamerican cities and has begun training three regional governments in chatbot design and implementation through the Public and Civic Innovation Program (PIIP) developed by SEGIB.

From a data governance perspective, Boti operates within strict local and international regulatory frameworks, leveraging Azure's native security capabilities. The system implements human-in-the-loop processes for oversight and to correct model outputs. However, an independent study by Derechos Digitales (2024) examined the chatbot's NLP and data handling practices, and no independent privacy audit of the GenAI components has been publicly documented. The Auditoría General de la Ciudad de Buenos Aires (AGCBA) has also conducted an audit of the Boti system. Future development plans include voice assistant integration, SEO optimisation, automatic learning from user interactions, and expansion of GenAI capabilities beyond tourism to additional government service domains.

Classifications follow the DCI AI Hub Taxonomy. Hover over field labels for definitions.

Social Protection Functions

Implementation/delivery chain
Outreach/communications/sensitisation primaryAssessment of needs/conditions + enrolment Case management
SP Pillar (Primary) The social protection branch: social assistance, social insurance, or labour market programmes. Social assistance
Programme Name Buenos Aires City Citizen Services Platform (Boti WhatsApp Chatbot)
Programme Type The type of social protection programme, classified under social assistance, social insurance, or labour market programmes. View in glossary Other
System Level Where in the social protection system the AI is applied: policy level, programme design, or implementation/delivery chain. View in glossary Implementation/delivery chain
Automation Subtype For operational automation cases: (a) document processing and generative staff assistance, or (b) workload and resource forecasting. (a) Document processing and generative staff assistance
Programme Description Multi-service municipal chatbot operated by the Government of the City of Buenos Aires providing citizen access to social assistance programmes, public health services, administrative procedures, and urban services through WhatsApp, with generative AI capabilities powered by Microsoft Azure OpenAI (GPT-4o with RAG).
Implementation Type How the AI output is produced: Classical ML, Deep learning, Foundation model, or Hybrid. Affects validation, compute requirements, and governance profile. View in glossary Foundation model
Lifecycle Stage Current stage in the AI lifecycle, from problem identification through to monitoring, maintenance and decommissioning. View in glossary Integration and Deployment
Model Provenance Origin of the AI model: developed in-house, adapted from open-source, commercial/proprietary, or accessed via third-party API. View in glossary API-accessed third-party
Compute Environment Where the AI system runs: on-premise, government cloud, commercial cloud, or edge/device. View in glossary Commercial cloud
Compute Provider The specific cloud or infrastructure provider hosting the AI system. Microsoft Azure
Sovereignty Quadrant Classification of data and compute sovereignty: I (Sovereign), II (Federated/Hybrid), III (Cloud with safeguards), or IV (Shared Innovation Zone). View in glossary III — Compute-Intensive Cloud with safeguards
Data Residency Where the data used by the AI system is stored: domestic, regional, or international. View in glossary Not documented
Data Residency Detail Additional detail on the specific data hosting arrangements and jurisdictions. System uses Microsoft Azure OpenAI Services; no public source confirms whether data is processed within Argentina, regionally in Latin America, or internationally via Azure data centres
Cross-Border Transfer Whether data crosses national borders, and if so, whether documented safeguards are in place. View in glossary Not documented
Decision Criticality The rights impact of the decision the AI supports. High criticality requires HITL oversight; moderate requires HOTL; low may operate HOOTL. View in glossary Low
Human Oversight Type Level of human involvement: Human-in-the-Loop (active review), Human-on-the-Loop (monitoring), or Human-out-of-the-Loop (periodic audit). View in glossary HITL
Development Process Whether the AI system was developed fully in-house, through a mix of in-house and third-party, or fully by an external provider. View in glossary Mix of in-house and third-party
Highest Risk Category The most significant structural risk source identified: data, model, operational, governance, or market/sovereignty risks. View in glossary Market, sovereignty and industry structure risks
Risk Assessment Status Whether a formal risk assessment, informal assessment, or independent audit has been conducted for this system. Informal assessment

Impact Dimensions

Autonomy, human dignity and due process
  • Data minimisation controls
  • Human oversight protocol
CategorySensitivityCross-System LinkageAvailabilityKey Constraints
Administrative data from other sectorsPersonalLinks data across multiple systemsCurrently available and usedIntegrated with social assistance systems under Ministry of Human Development and Habitat for housing subsidy tracking, social programme information, and homelessness assistance requests
Unstructured and text-based contentNon-personalLinks data across multiple systemsCurrently available and usedCentralised government content repository used for RAG grounding; includes official tourism data from Ente de Turismo de la Ciudad; over 450 topics across 70+ government system integrations

Botmaker (n.d.) 'Boti, the bot of the City of Buenos Aires', Botmaker Client Case Study. Available at: https://botmaker.com/en/clients/boti-caba-bot (Accessed: 25 March 2026).

View source Government website / press release

Gobierno de la Ciudad Autónoma de Buenos Aires (2022) 'Boti se incorpora como nuevo canal para realizar consultas y trámites de atención social', Buenos Aires Ciudad, 18 April. Available at: https://buenosaires.gob.ar/desarrollohumanoyhabitat/atencion-social/noticias/boti-se-incorpora-como-nuevo-canal-para-realizar (Accessed: 25 March 2026).

View source Government website / press release

Microsoft (2024) 'Buenos Aires City: How generative AI is revolutionizing the lives of millions with Azure OpenAI Services', Microsoft Customer Stories. Available at: https://www.microsoft.com/en/customers/story/21596-government-of-the-city-of-buenos-aires-azure-open-ai-service (Accessed: 25 March 2026).

View source Government website / press release

Microsoft News Center Latinoamérica (2024) 'Buenos Aires: una ciudad cada vez más inteligente con Boti usando IA', Microsoft News Center, 13 August. Available at: https://news.microsoft.com/es-xl/buenos-aires-una-ciudad-cada-vez-mas-inteligente-con-boti-usando-ia/ (Accessed: 25 March 2026).

View source News article / media

OECD Observatory of Public Sector Innovation (2023) 'Boti, the City's WhatsApp!', OPSI Innovation Case Study. Available at: https://oecd-opsi.org/innovations/boti-the-citys-whatsapp/ (Accessed: 25 March 2026).

View source Report (multilateral / development partner)

Windows Forum (2024) 'Public AI Chatbots in Madrid and Buenos Aires Backed by Azure OpenAI', Windows Forum, 2024. Available at: https://windowsforum.com/threads/public-ai-chatbots-in-madrid-and-buenos-aires-backed-by-azure-openai.394536/ (Accessed: 25 March 2026).

View source News article / media
Deployment Status How far the system has progressed into real-world operational use, from concept/exploration through to scaled and institutionalised. View in glossary Full Production Deployment
Year Initiated The year the AI system was first initiated or development began. 2019
Scale / Coverage The scale and geographic or population coverage of the deployment. Approximately 2 million queries per month without human intervention; over 153 million cumulative conversations since 2019 launch; 450+ topics; 70+ system integrations; 82% of queries resolved without human intervention; serves residents and visitors of Buenos Aires (population ~3 million, metropolitan area ~15 million)
Funding Source The source(s) of funding for the AI system development and deployment. Government of the City of Buenos Aires municipal budget; Microsoft partnership (Azure OpenAI Services)
Technical Partners External technology vendors, academic partners, or development partners involved. Microsoft (Azure OpenAI Services, GPT-4o); Botmaker (original chatbot platform); Pi Data Strategy & Consulting (GenAI integration consulting)
Outcomes / Results 2 million queries per month handled without human intervention; 82% automated resolution rate; 50% reduction in operational burden for municipal teams; 90% average monthly satisfaction rate; over 11 million vaccination appointments facilitated during COVID-19 campaign; near-100% monthly service availability; selected to train 3 Iberoamerican cities in chatbot implementation; Best Corporate Chatbot 2019 (Planeta Chatbot Awards); recognised by Facebook as international COVID-19 vaccination solution example
Challenges WhatsApp platform constraint of 200 messages per second caused bottlenecks during COVID-19 peak demand (over 400 messages per second for test results), requiring overflow to Facebook Messenger and web channels; ensuring GenAI system respects Argentine dialect, inclusive language, and government tone; centralising all government information into single repository for RAG grounding; maintaining accuracy and preventing hallucination in citizen-facing responses on government services; no independent privacy audit of GenAI components publicly documented; data residency and cross-border transfer status for Azure-processed queries undocumented

How to Cite

DCI AI Hub (2026). 'Buenos Aires Boti GenAI Chatbot (WhatsApp Citizen Services with Azure OpenAI)', AI Hub AI Tracker, case ARG-001. Digital Convergence Initiative. Available at: https://socialprotectionai.org/use-case/ARG-001 [Accessed: 1 April 2026].

Change History

Created 30 Mar 2026, 08:37
by v2-import (import)