Seattle's Innovation & Performance team and SDCI ran real residential permit applications through CivCheck's AI pre-screening and compared results with staff review: completeness checks were 87% accurate and design compliance checks 92% accurate, but incomplete check coverage limited cycle savings, so the city recommended a completeness-only production pilot (reported by City of Seattle)
Construction permit applications usually take several review cycles, each adding a month or more, which adds cost and delay to housing projects (reported by City of Seattle)
Applicant for a small residential project (townhouse, single-family home, ADU) runs the application through CivCheck before submitting
AI checks document completeness and design compliance against City requirements and returns feedback
Applicant fixes flagged errors, then submits into the normal SDCI review
In the study, CivCheck findings were compared with SDCI staff reviews of the same applications and impact on review cycles was modeled
Human-in-the-loop: City review is unchanged; AI only gives applicants pre-submission feedback. Staff reviewed the same applications to benchmark the AI.
Partial coverage creating false confidence for applicants
Hard-to-automate subjective code requirements
Limitations
Proof-of-value study run March–October 2025, not a production deployment; cycle-time impact was modeled rather than measured live (reported by City of Seattle)
City notes thousands of code requirements make compliance automation hard and says simplifying its own codes is needed
Same vendor as the Honolulu record; Seattle's independent findings are more cautious than the vendor case study
Completeness checks 87% accurate and design compliance checks 92% accurate versus SDCI staff review Reported by City of Seattle · Source (innovation-hub.seattle.gov)
City found check coverage matters as much as accuracy—pre-screening must catch every correction to eliminate a review cycle—and recommended a production pilot for completeness pre-screening only Reported by City of Seattle · Source (innovation-hub.seattle.gov)
Amgen built Catalyst Copilot in Copilot Studio in ~6 weeks so R&D staff can ask natural-language questions over curated drug-development knowledge (reported by Microsoft / Amgen)
Multi-agent Bedrock system lets plant engineers query IoT data in natural language for bottleneck root cause analysis across 250+ curing presses (reported by Apollo Tyres / AWS)
Bosch Bamberg plant operators use the Shopfloor Agent to diagnose machine errors from stored knowledge and restart production faster (reported by Bosch)
Specialty insurer CFC launched Lane Assist, a live pilot of agentic underwriting in its Cyber team that takes low-complexity submissions from email to quote recommendation in seconds; every recommendation is reviewed and approved by an underwriter before issue (reported by CFC)
Honolulu's Department of Planning and Permitting uses Clariti's CivCheck guided AI plan review to pre-screen residential permit applications for completeness and code compliance before staff review; average time to permit decision fell from 73 to 32.5 days on a small Q1 2026 comparison sample (reported by Clariti / Honolulu DPP)