BGD-003

"Smart Sathi" Chatbot – 333 National Helpline

Download PDF
Bangladesh South Asia Lower middle income Pilot / Controlled Trial Phase Confirmed

Aspire to Innovate (a2i), ICT Division, Cabinet Division, Government of Bangladesh

At a Glance

What it does Perception and extraction from unstructured inputs — User communication and interaction
Who runs it Aspire to Innovate (a2i), ICT Division, Cabinet Division, Government of Bangladesh
Programme National Helpline 333
Confidence Confirmed
Deployment Status Pilot / Controlled Trial Phase
Key Risks Model-related risks
Key Outcomes Over 100 million calls handled for government information/services since 2018 inception; 164 million emergency assistance calls; 150,000 social issue prevention calls.
Source Quality 6 sources — Government website / press release, News article / media

The "Smart Sathi" (SmartSathi) chatbot is an artificial intelligence-powered conversational assistant integrated into Bangladesh's National Helpline 333, a centralized voice-based platform for government information dissemination, grievance redress, and service application assistance. The helpline was originally launched on April 12, 2018 by Aspire to Innovate (a2i) under the Information and Communication Technology (ICT) Division of the Government of Bangladesh and operates under the Cabinet Division. The SmartSathi AI component was inaugurated on October 18, 2023 on a trial basis by the then-Prime Minister, to provide information related to government services through voice, non-voice, and AI conversational bot channels.

The 333 helpline operates as a single voice access point for all government information and grievance redressal, available 24 hours a day, 7 days a week. Citizens can dial 333 from any mobile phone or landline, or use the number 09666789333 from landlines and international connections. The platform is designed to ensure that every citizen, regardless of internet access or digital literacy, has round-the-clock access to accurate government information, essential services, and mechanisms to report grievances. This is particularly significant in Bangladesh where more than 60 percent of the population lacks reliable internet connectivity, making the voice-based channel a critical tool for bridging the digital divide.

The SmartSathi conversational bot was specifically designed to help citizens access government services in a simplified way with a minimal human workforce, while also reducing the burden of unnecessary prank calls on human operators. The system provides automated responses through natural language processing capabilities in Bangla, enabling citizens to receive information on public services, social benefits, allowances, and to submit grievances. The bot operates across multiple channels, including interactive voice response (IVR), conversational bot interfaces, and text-based chat. There are plans to expand the SmartSathi service to cover all information available through the National Portal and myGov platform, encompassing 1,898 services across 35 government ministries.

The range of services accessible through the 333 helpline and SmartSathi includes government service information and guidance, social issue complaint registration and resolution (covering child marriage, violence against women and children, environmental pollution, drug abuse, food adulteration, and eve teasing), specialist doctor consultations, government land services, service-related complaint reporting, cyber security assistance, legal support for women and children, and emergency and disaster assistance. The helpline also provides information on social safety net programmes including government allowances and food assistance. During Cyclone Mocha, the system received 1.4 million calls and processed over 34,000 disaster-specific emergency assistance requests, demonstrating its role as a critical emergency response channel.

Since its inception, the 333 helpline has handled over 100 million calls for government information and services, approximately 150,000 calls for social issue prevention, and 164 million calls for emergency assistance. The helpline receives approximately 25,000 calls each day, of which 8,000 to 10,000 calls are served directly through the helpline operators. The AI-enabled voice bots handle a portion of these calls through automated responses, with complex or escalated queries routed to live human agents for review and resolution.

The technology stack incorporates AI-enabled voice bots, omnichannel communication capabilities, and multilingual support. The system uses Natural Language Processing (NLP) with Bangla language models as its core AI technique. The chatbot infrastructure is reported to operate on government cloud infrastructure under the National Data Center managed by the Bangladesh Computer Council, though this specific claim requires further verification. The system includes Chatwoot integration for live chat support and an accessibility widget for users with disabilities.

The human oversight model follows a human-in-the-loop (HITL) approach where live agents review escalated or complex queries for accuracy and citizen protection. The decision criticality is assessed as low, as the system provides advisory and informational responses with no direct impact on benefit eligibility or entitlements. The SmartSathi bot serves as a first-line information and triage tool, with human operators available for cases that exceed the bot's capabilities or require nuanced judgement.

The system was jointly developed by a2i and the ICT Division, with UNDP providing institutional support to the broader a2i programme. The Bangladesh Telecommunication Regulatory Commission (BTRC) has provided funding for the related Sathi virtual assistant app component. The system is aligned with the Smart Bangladesh Vision 2041 and operates within the framework of the National Digital Governance Architecture. References to a2i's AI Ethics and Governance Guidelines (2023 draft) as a regulatory framework exist, though independent verification of these guidelines and their specific application to SmartSathi has not been confirmed through the available sources.

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

Social Protection Functions

Implementation/delivery chain
Outreach/communications/sensitisation primaryAccountability mechanisms
SP Pillar (Primary) The social protection branch: social assistance, social insurance, or labour market programmes. Social assistance
Programme Name National Helpline 333
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
Programme Description Bangladesh's centralized national helpline providing 24/7 voice-based access to government information, grievance redressal, and service application assistance across all government ministries, operated by Aspire to Innovate (a2i) under the ICT Division and Cabinet Division.
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 Deep learning
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 Not documented
Compute Environment Where the AI system runs: on-premise, government cloud, commercial cloud, or edge/device. View in glossary National/government cloud
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 II — Federated/Hybrid Governance
Data Residency Where the data used by the AI system is stored: domestic, regional, or international. View in glossary Not documented
Cross-Border Transfer Whether data crosses national borders, and if so, whether documented safeguards are in place. View in glossary Not documented
Is Agentic Whether the system autonomously plans and executes multi-step workflows, selecting tools and chaining actions with limited human intervention. View in glossary Partial
Agentic Pipeline Description of the chained workflow steps in the agentic pipeline. SmartSathi bot handles initial user queries via IVR, voice, and chat channels; escalates complex or unresolved queries to live human agents for review and resolution.
Agentic Autonomy Degree of autonomy: fully autonomous, semi-autonomous (human checkpoints), or supervised (human approval at each step). Semi-autonomous
Override Points Where in the pipeline human review or override is triggered. Live agents review escalated or complex queries; human operators available for cases exceeding bot capabilities.
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 Model-related risks
Risk Assessment Status Whether a formal risk assessment, informal assessment, or independent audit has been conducted for this system. Not assessed

Impact Dimensions

Autonomy, human dignity and due process
Equality, non-discrimination, fairness and inclusion
Systemic and societal
  • Grievance mechanism
  • Human oversight protocol
CategorySensitivityCross-System LinkageAvailabilityKey Constraints
Unstructured and text-based contentPersonalSingle source (no linkage)Currently available and usedText transcripts and anonymised interaction metadata from citizen helpline calls; Bangla language content requiring specialised NLP processing.

a2i (2025) National Helpline 333. Dhaka: Aspire to Innovate (a2i), ICT Division, Government of Bangladesh. Available at: https://a2i.gov.bd/site/page/ccdeba3d-0db9-4640-ab28-e5a7cf96019b/- (Accessed: 23 March 2026).

View source Government website / press release

BSS (2023) '333's SmartSathi' to further simplify receiving govt services. Dhaka: Bangladesh Sangbad Sangstha (BSS). Available at: https://www.bssnews.net/journey-of-development/162349 (Accessed: 23 March 2026).

View source News article / media

333.gov.bd (2025) National Helpline 333 – Official Website. Dhaka: Government of Bangladesh. Available at: https://333.gov.bd/ (Accessed: 23 March 2026).

View source Government website / press release

The Business Standard (2024) 'Emergency Helpline 333 to be launched for price complaints: Palak'. Dhaka: The Business Standard. Available at: https://www.tbsnews.net/bangladesh/emergency-helpline-333-be-launched-price-complaints-palak-775414 (Accessed: 23 March 2026).

View source News article / media

The Daily Star (2023) 'Government plans to launch "Saathi", a Bangladeshi version of Siri'. Dhaka: The Daily Star. Available at: https://www.thedailystar.net/tech-startup/news/government-plans-launch-saathi-bangladeshi-version-siri-3492681 (Accessed: 23 March 2026).

View source News article / media

The Financial Express (2023) 'a2i's yearlong initiatives build foundation of Smart Bangladesh'. Dhaka: The Financial Express. Available at: https://thefinancialexpress.com.bd/national/a2is-yearlong-initiatives-build-foundation-of-smart-bangladesh (Accessed: 23 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 Pilot / Controlled Trial Phase
Year Initiated The year the AI system was first initiated or development began. 2023
Scale / Coverage The scale and geographic or population coverage of the deployment. National scope (333 helpline); SmartSathi AI bot launched on trial basis October 2023. Helpline receives ~25,000 calls/day; 100+ million total calls since 2018 inception.
Funding Source The source(s) of funding for the AI system development and deployment. Government of Bangladesh (ICT Division); BTRC funding for Sathi app component; UNDP institutional support to a2i programme.
Technical Partners External technology vendors, academic partners, or development partners involved. a2i and ICT Division jointly developed the system. UNDP provides institutional support. BTRC funded the related Sathi virtual assistant app.
Outcomes / Results Over 100 million calls handled for government information/services since 2018 inception; 164 million emergency assistance calls; 150,000 social issue prevention calls. During Cyclone Mocha: 1.4 million calls received, 34,000+ emergency assistance requests processed. Approximately 25,000 calls received daily with 8,000-10,000 served directly. Expanded reach to rural users through Bangla voice and text channels, bridging the digital divide for the 60%+ population lacking reliable internet.

How to Cite

DCI AI Hub (2026). '"Smart Sathi" Chatbot – 333 National Helpline', AI Hub AI Tracker, case BGD-003. Digital Convergence Initiative. Available at: https://socialprotectionai.org/use-case/BGD-003 [Accessed: 1 April 2026].

Change History

Updated 1 Apr 2026, 08:11
by system (system)
Created 30 Mar 2026, 08:38
by v2-import (import)