From the Citizen Engagement & Services edition of November 2, 2026
Eagle Mountain, Utah launches Emmie, a Citibot AI assistant for web chat and text that takes requests in 70+ languages with handoff to staff
Eagle Mountain City, Utah (vendor: Citibot) · City constituent services and 311-style requests · Eagle Mountain, Utah, United States
- Publisher
- Eagle Mountain City
- Original publication
- 2026-11-02
- Source retrieved
- 2026-11-03
- Event date
- 2026-11-02
What happened
On Nov. 2 the city of more than 80,000 residents announced Emmie, an AI assistant built with Citibot, available by web chat and by text message. The city says residents can ask questions, report issues and submit requests in more than 70 languages, and that Emmie can pass requests to City staff when it cannot help. The post says Emmie is not for emergencies. No usage, accuracy, cost or translation-quality data is given.
Why it matters
A fresh example of a small, fast-growing city buying an off-the-shelf AI front door that combines information, request intake, SMS and multilingual access. Citibot is the same vendor whose Denver deployment was rejected by council, so buyers have comparable cases to ask about.
Evidence and measured results
City announcement only; operator claims; no metrics, evaluation or contract terms.
Limitations and uncertainty
Launch announcement; no outcomes; translation accuracy across 70 languages unverified.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-11-03; 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 want after-hours answers and request intake without calling City Hall.
- Stakeholders
- City manager or communications director
- Public works and request owners
- City IT
- Language access or community outreach staff
- Discovery questions
- Which request types should the assistant take, and where do they land?
- Which languages do residents use most?
- How will handoffs and unresolved questions be measured?
- Value hypothesis
- An AI front door may handle routine questions after hours; no results are reported yet.
- Potential engagement
- A 90-day evaluation plan with baseline call and request data and translation spot checks.
- Unsupported claims
- Fewer calls or staff hours
- Accurate translation in 70 languages
- Resident satisfaction gains
Pre-sales engineering
Role takeaway
- Fit
- Fits small and mid-size cities without a 311 center.
- Architecture and integration
- Hosted chat and SMS assistant grounded in city web content, passing requests and handoffs to staff.
- Prerequisites
- Current city web content
- Request routing rules
- Staff handoff queue
- Constraints
- Translation quality unverified
- SMS carrier and consent rules
- Not for emergencies
- Security
- Set retention limits for chat and text logs and limit vendor reuse of resident data.
- Proof of value
- Test top questions for accuracy
- Native-speaker check of translated answers
- Screen-reader test of web chat
Delivery
Role takeaway
- Work
- Run and measure the assistant after launch.
- Dependencies
- Vendor configuration
- Department content owners
- Request system integration
- Ownership
- City communications office, with departments owning their answers.
- Skills and adoption
- Staff need training on handling handoffs and updating content.
- Governance checkpoints
- 30-day review of unresolved questions
- Translation spot check
- Annual contract and data-use review
- Acceptance criteria
- Proposed: every handoff answered by staff within a set time
- Proposed: monthly accuracy sample above an agreed threshold
- Risks
- Wrong answers presented as official
- Poor translations
- Requests lost between assistant and staff
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?
Web chat and SMS channels feed request intake; the integration with the city's work-order or request system is not described.
Governance
Who approves, reviews and stays accountable for outcomes?
A human handoff is stated; there is no published AI use policy, accuracy testing or disclosure of translation method in the post.
Security and privacy
What data, permissions and controls need testing?
Text-message intake collects phone numbers and request details; retention and vendor data use are not stated.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Multilingual and SMS access may widen reach, but machine translation quality and accessibility for screen-reader users are not addressed.
Procurement
What should contracts, pricing and exit terms secure?
Buyers should ask the vendor for reference results from comparable cities, including ones that ended contracts.
Operating model
Which teams own the service once it runs?
Staff must monitor handoffs and keep the knowledge base current.
Publication history
Read preserved resource versions (JSON)Stable resource ID: eagle-mountain-ut-emmie-citibot-ai-assistant-20261102