From the Citizen Engagement & Services edition of November 10, 2026
Porto Alegre puts 'Sol', a generative AI agent, on the WhatsApp channel of its 156 citizen line from Nov. 11, with handoff to human agents and voice and portal phases to follow
Prefeitura de Porto Alegre (Central do Cidadão 156; Secretaria Municipal de Transparência e Controladoria) · Municipal 311-style citizen contact centre · Porto Alegre, Rio Grande do Sul, Brazil
- Publisher
- Prefeitura de Porto Alegre (news release)
- Original publication
- 2026-11-10
- Source retrieved
- 2026-11-11
- Event date
- 2026-11-11
What happened
On Nov. 10, 2026 Porto Alegre presented AI-based service on the WhatsApp channel of its Central do Cidadão 156, available from Nov. 11. Residents write or send audio in their own words, with no menus, and receive answers based on official city content; when Sol finds the requested service it offers to open the request inside the conversation, and residents can follow up on existing service protocols. Cases that need a person are transferred to a 156 agent along with the conversation history, and users can rate the service. The city says 156 has handled more than 6 million interactions in about two and a half years, and that later phases bring Sol to voice calls, a new web service portal and an 'intelligent map' of demand for managers.
Why it matters
A direct analogue of a U.S. 311 adding a generative agent on a messaging channel residents already use, with request intake, status lookup and warm handoff in scope from day one. Launched the day of this edition, so it is a current signal; outcomes are not yet known.
Evidence and measured results
City launch announcement with stated design features and a cumulative volume figure; no accuracy testing, containment target, vendor, model or evaluation plan is published.
Limitations and uncertainty
Launch-day announcement; vendor and model unnamed; no performance data for the AI channel; Brazilian context (WhatsApp ubiquity, LGPD) limits direct transfer to U.S. cities.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-11-10; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Residents waiting on a contact centre for routine questions and request status that could be handled in a messaging app.
- Stakeholders
- 311 / contact-centre director
- Transparency or CX office
- Case-system owner
- Data-protection officer
- Discovery questions
- Which channels do residents use most for 311 today?
- Can an agent safely create and read service requests?
- How will handoff quality and ratings be reported?
- Value hypothesis
- A messaging-channel agent with request intake and warm handoff may shorten routine contacts; no outcome data yet supports a size.
- Potential engagement
- A scoped WhatsApp/SMS agent pilot covering three high-volume request types plus status lookup and handoff.
- Unsupported claims
- Guaranteed containment rate
- Staff reductions
- Accuracy (not published)
Pre-sales engineering
Role takeaway
- Fit
- Fits 311 programmes with a case system that exposes APIs and residents who use messaging apps.
- Architecture and integration
- Messaging-platform agent grounded in official content, with audio transcription, case creation, status lookup and history handoff to agents.
- Prerequisites
- Case-management API
- Identity matching for status lookups
- Curated official knowledge base
- Constraints
- Third-party messaging platform terms
- Brazil-specific data-protection regime
- Vendor and model unknown
- Security
- Define authentication for status lookups, retention of chats and audio, and platform data-sharing terms.
- Proof of value
- Request-creation accuracy on a test set
- Status-lookup correctness
- Completeness of history passed at handoff
Delivery
Role takeaway
- Work
- Stand up the messaging agent, connect it to the case system and run it alongside human agents.
- Dependencies
- Case-management system
- Messaging-platform account
- Contact-centre staff
- Ownership
- Contact-centre (156) leadership with the transparency and control secretariat.
- Skills and adoption
- Train agents to take over mid-conversation and content staff to maintain official answers.
- Governance checkpoints
- Pre-launch routing test sign-off
- Weekly conversation sample review
- Go/no-go gate before voice phase
- Acceptance criteria
- Routing accuracy above an agreed threshold
- History attached to 100% of handoffs
- Monthly user ratings published
- Risks
- Wrong requests created in the case system
- Status disclosed to the wrong person
- Channel rollout outpacing evaluation
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Messaging-channel agent (WhatsApp) answering from official city content, accepting audio, opening service requests in-conversation and reading protocol status, so it must integrate with the 156 case system and pass history to agents.
Governance
Who approves, reviews and stays accountable for outcomes?
The release names the transparency and control secretariat as owner but does not describe answer review, error correction or how the city will measure accuracy; Porto Alegre's municipal AI law (archived Oct. 2026) would be the governing frame, interpretation only.
Security and privacy
What data, permissions and controls need testing?
Status lookup and request intake over WhatsApp imply handling of personal data and identifiers on a third-party platform; retention and data-protection terms are not described.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Audio input may help residents with low literacy; human agents remain for complex cases and receive conversation history, so residents need not repeat themselves.
Procurement
What should contracts, pricing and exit terms secure?
No contract, vendor or cost is stated; buyers would need to ask how a messaging-platform agent is procured and priced.
Operating model
Which teams own the service once it runs?
First phase of a staged plan (WhatsApp, then voice, then portal and demand map), positioning one AI persona across all 156 channels.
Publication history
Read preserved resource versions (JSON)Stable resource ID: porto-alegre-156-sol-whatsapp-ai-agent-launch-20261110