|
Field
Abstract
|
Before
This paper examines how artificial intelligence can enhance administrative efficiency in public service delivery, focusing on the processing of building permit applications. We evaluate an AI-assisted system specifically designed and implemented to experimentally assess its impact on bureaucratic performance. Using a field experiment in Chilean municipalities, the platform integrates AI tools into the review process to support technical assessment while preserving regulatory discretion. The analysis focuses on how AI adoption affects processing times, workflow efficiency, and administrative congestion—key sources of delay that generate economic costs for the construction sector and broader welfare losses. By reducing waiting times within a highly regulated approval process, the intervention illustrates how purpose-built AI systems can improve productivity in highly congested and bureaucratic institutional environments. Finally, the study explores heterogeneous effects using reviewers’ pre-intervention administrative histories to assess how baseline performance conditions shape the impact of AI assistance.
|
After
This paper examines how artificial intelligence can enhance administrative efficiency in public service delivery, focusing on the processing of building permit applications. We evaluate an AI-assisted system specifically designed and implemented to experimentally assess its impact on bureaucratic performance. Using a field experiment in Chile involving architects reviewing building permit applications under conditions that mimic the work of municipal reviewers, the platform integrates AI tools into the review process to support technical assessment while preserving regulatory discretion. The analysis focuses on how AI adoption affects processing times, workflow efficiency, and administrative congestion—key sources of delay that generate economic costs for the construction sector and broader welfare losses. By reducing waiting times within a highly regulated approval process, the intervention illustrates how purpose-built AI systems can improve productivity in highly congested and bureaucratic institutional environments. Finally, the study explores heterogeneous effects using reviewers’ features to assess how baseline performance conditions shape the impact of AI assistance.
|
|
Field
Trial Start Date
|
Before
January 02, 2026
|
After
October 05, 2026
|
|
Field
Trial End Date
|
Before
December 31, 2026
|
After
March 31, 2027
|
|
Field
Last Published
|
Before
January 06, 2026 07:13 AM
|
After
October 08, 2026 12:27 PM
|
|
Field
Intervention (Public)
|
Before
The intervention introduces an AI-assisted decision-support system into the administrative review of building permit applications. The system is designed to support public officials by structuring regulatory checks, organizing documentation, and facilitating the technical assessment of compliance with existing building codes. Applications are processed either with or without AI assistance, allowing for a comparison of review times and workflow efficiency. The intervention aims to reduce administrative delays and improve productivity in a highly regulated and congestion-prone public service environment.
|
After
The intervention introduces an AI-assisted review platform into the first technical review of minor works permit applications. In a controlled laboratory setting, six architects review the same 34 real applications from two Chilean municipalities: each application is reviewed by three of them with AI assistance and by the other three without it. This design allows a comparison of review times, the quality of the resulting observation reports, and the consistency of assessments across reviewers. The intervention aims to reduce administrative delays and improve productivity without compromising the quality of regulatory review.
|
|
Field
Intervention Start Date
|
Before
January 15, 2026
|
After
October 07, 2026
|
|
Field
Intervention End Date
|
Before
April 01, 2026
|
After
November 23, 2026
|
|
Field
Primary Outcomes (End Points)
|
Before
The primary outcome of interest is the time required to complete the technical review of building permit applications, measured from the date of formal submission to the completion of the initial administrative assessment. This outcome captures the core dimension of administrative efficiency and directly reflects delays experienced by applicants and the broader construction sector. Processing time is measured using administrative system records and allows for precise comparison across treatment and control conditions. The primary analysis focuses on whether access to AI-assisted decision support reduces review duration relative to standard manual procedures. By centering on processing time, the study evaluates whether AI tools can meaningfully improve the speed of service delivery in a highly regulated and capacity-constrained public sector environment, while maintaining existing institutional rules and professional discretion.
|
After
The primary outcome of interest is the effective time required to complete the first technical review of a minor works permit application, measured in minutes from the moment the reviewer opens the application until the observation report is submitted. It excludes pauses and, in the treatment arm, includes the time needed to load the application into the AI platform. This outcome captures a key dimension of administrative efficiency and directly reflects the delays experienced by applicants and the construction sector. Review time is recorded with the reviewer's stopwatch log and reconciled daily with the start and end times recorded by the coordinator, allowing precise comparisons between treatment and control conditions. The primary analysis estimates the effect of AI assistance on the logarithm of review time, with reviewer and application fixed effects and a one-sided test, and is interpreted jointly with the quality and consistency of the resulting reports. By focusing on review time without neglecting quality, the study evaluates whether AI tools can meaningfully accelerate permit review while preserving existing regulations and professional judgment.
|
|
Field
Experimental Design (Public)
|
Before
The study implements a randomized controlled trial (RCT) to evaluate the causal impact of AI-assisted decision support on the processing of building permit applications in two Chilean municipalities. The intervention is embedded within routine administrative operations and compares standard manual review procedures with an AI-assisted workflow. Assignment to treatment and control is conducted using a completely randomized design within predefined strata defined by reviewer and permit type. This ensures balance between treatment and control groups across key dimensions of administrative heterogeneity. The design allows for credible causal inference on the effects of AI assistance on processing times and administrative efficiency, while preserving existing institutional rules and decision authority.
|
After
The study implements a randomized controlled trial (RCT) in a laboratory setting to evaluate the causal impact of AI assistance on the first technical review of minor works permit applications from two Chilean municipalities. It compares review as currently conducted in a municipal works department, including any standard support such as other AI tools, with review using the AI platform as the preferred support. Assignment follows a crossed design with complementary pairing: each application is reviewed by all six reviewers, three with AI assistance and three without, and each reviewer handles half of the applications under each condition. This ensures exact balance across applications and reviewers, the two main sources of heterogeneity in review times. The design allows credible causal inference on the effects of AI assistance on review time, quality and consistency, while reviewers retain full authority over each observation report under existing regulations.
|
|
Field
Randomization Method
|
Before
Randomization is implemented through a computerized procedure designed to ensure a Completely Randomized Design (CRD) within predefined strata. Specifically, building permit applications are randomly assigned to treatment or control groups within strata defined by reviewer and permit type. Within each stratum, assignment is fully random and independent, ensuring balanced allocation between treatment and control conditions while preserving the simplicity and transparency of a CRD framework. The randomization procedure is programmed to follow established best practices in experimental design, ensuring reproducibility, balance, and protection against discretionary assignment or manipulation. This approach supports unbiased estimation of treatment effects under standard assumptions of randomized experiments.
|
After
Randomization is implemented through a computerized procedure designed to ensure exact balance through complementary pairing. Specifically, the 34 applications are grouped into 17 fixed pairs, in an order set before the draw. For each pair, the procedure randomly selects which application comes first and one of the 20 possible subsets of three reviewers who review it with AI assistance; the complementary subset reviews the second application with AI assistance. Each reviewer's review order is then drawn as an independent random permutation, and each reviewer-application combination receives a unique blind code. The procedure uses a pseudorandom generator with fixed seeds and is replicated bit by bit by two independent implementations, ensuring reproducibility, balance and protection against discretionary assignment or manipulation. This approach supports unbiased estimation of treatment effects under standard assumptions of randomized experiments.
|
|
Field
Randomization Unit
|
Before
The unit of randomization is the individual building permit application. Each application is independently assigned to treatment or control within its corresponding stratum. Reviewers may therefore evaluate both treated and control applications, enabling efficient estimation of treatment effects while accounting for reviewer-specific heterogeneity.
|
After
The unit of randomization is the individual review, defined as a reviewer-application combination. Each application is reviewed by all six reviewers, and treatment is assigned at the review level: three reviews of each application are conducted with AI assistance and three without. Every reviewer therefore evaluates both treated and control applications, and every application is reviewed under both conditions, enabling efficient estimation of treatment effects while accounting for reviewer-specific and application-specific heterogeneity.
|
|
Field
Planned Number of Observations
|
Before
The planned sample consists of approximately 240–360 building permit applications. Each municipality processes on average 60 eligible applications per month, and data collection is expected to span 2 to 3 months. The unit of observation is the individual permit application. Applications are distributed across reviewers and permit types, generating variation both within and across strata defined by reviewer × permit type.
|
After
The sample consists of 34 real minor works applications (14 from Renca and 20 from Rancagua), each reviewed by six reviewers, for a total of 204 reviews. Applications were selected among first-review submissions recorded in the AI platform and cover the three types of minor works: extensions of up to 100 m² (18), modifications without structural changes (13) and social or low-cost housing extensions (3). Data collection takes place between October 7 and 23, 2026 (12 working days). The unit of observation is the individual review (reviewer × application), generating variation both within and across applications and reviewers.
|
|
Field
Sample size (or number of clusters) by treatment arms
|
Before
Applications are randomly assigned to treatment and control groups using a completely randomized design within each reviewer × permit type stratum. Within each stratum, approximately 50% of applications are assigned to the AI-assisted treatment and 50% to the control condition. This implies an expected allocation of roughly 120–180 treated and 120–180 control applications overall, depending on realized inflow. This stratified CRD ensures balance across both reviewer-level and permit-type characteristics.
|
After
Reviews are assigned to treatment and control groups through complementary pairing within the 34 × 6 application × reviewer design. Each application receives exactly three AI-assisted and three control reviews, and each reviewer conducts exactly 17 reviews in each condition. This implies an allocation of 102 treated and 102 control reviews, fixed by design rather than dependent on realized inflow. This balanced design ensures that differences in reviewer speed and application difficulty are absorbed by reviewer and application fixed effects, so the treatment effect is estimated from within-reviewer and within-application contrasts.
|
|
Field
Power calculation: Minimum Detectable Effect Size for Main Outcomes
|
Before
|
After
Unit: individual review (reviewer × application); N = 204 reviews (34 applications × 6 reviewers). Main outcome: log effective review time, with a residual standard deviation of 0.61 log points, calibrated with minor works review times from a previous field pilot. With reviewer and application fixed effects, the crossed design yields a standard error of 0.085 for the treatment effect (σ·√(2/3N)). For a one-sided test at 5% with 80% power, the MDE is 0.21 log points (0.35 SD), equivalent to a 19.1% reduction in review time (about 13 minutes over the pilot mean of 69 minutes). Power to detect a 20% reduction is 83%; across the 90% confidence interval for σ (0.49–0.83), the MDE ranges from 15.7% to 25.1%. Inference will use randomization inference and standard errors clustered by reviewer and application.
|
|
Field
Intervention (Hidden)
|
Before
The study implements a randomized controlled trial embedded within the daily operations of municipal building departments. A purpose-built digital platform was developed to integrate artificial intelligence into the technical review of building permit applications. The intervention targets professional reviewers working in municipal planning and construction offices. Incoming applications are randomly assigned—within predefined strata based on reviewer and permit characteristics—to either a treatment group, where reviewers use AI-assisted tools, or a control group, where reviews are conducted using standard manual procedures.
The AI system supports reviewers by organizing documentation, flagging relevant regulatory provisions, and structuring compliance checks, while leaving all substantive judgments and final decisions entirely to human officials. The intervention operates fully within existing legal and institutional frameworks. Rich administrative data allow measurement of processing times, reviewer workload, and workflow congestion. This design enables credible causal inference on the effects of AI assistance on administrative efficiency and service delivery under real-world bureaucratic conditions.
|
After
The study implements a randomized controlled trial in which architects review real minor works permit applications under laboratory conditions that replicate the work of municipal building departments. An AI platform developed to support the technical review of permit applications is integrated into the review workflow. The intervention targets professional reviewers: six architects with experience in minor works review 34 real applications from two Chilean municipalities, and each application is reviewed by all six. Through complementary pairing, three reviewers assess each application with AI assistance and three without it, so every reviewer handles half of the applications in each condition; in both arms reviewers may use any standard support, including other AI tools.
The AI system supports reviewers by organizing documentation, giving direct access to the applicable regulations and structuring compliance checks, while all substantive judgments and the final observation report remain entirely with the human reviewer. The intervention operates fully within existing legal and institutional frameworks. Detailed records allow measurement of review time, the quality of each observation report under blind assessment, and the consistency of assessments across reviewers. This design enables credible causal inference on the effects of AI assistance on the efficiency and quality of permit review under conditions that simulate real-world bureaucratic work.
|
|
Field
Secondary Outcomes (End Points)
|
Before
Secondary outcomes examine how AI assistance affects patterns of congestion, workload distribution, and heterogeneity in performance across reviewers and permit types. These include measures of queue length and backlog accumulation, the number of applications processed per reviewer over time, and variation in processing times across case complexity categories. Additional outcomes assess whether AI support differentially benefits certain types of permits or reviewers, shedding light on where productivity gains are most pronounced. Together, these measures allow the analysis to move beyond average effects and explore how AI reshapes administrative dynamics, workflow organization, and bottlenecks within the permitting system. All secondary outcomes are constructed from administrative records generated during routine operations, ensuring consistency and comparability across treatment and control groups.
|
After
Secondary outcomes examine how AI assistance affects the quality and consistency of reviews, and how its effects vary across reviewers and permit types. Quality is measured by comparing each observation report, under blind assessment, with a reference report: the share of relevant and of substantive observations detected, observations without legal basis, and report precision. Consistency is measured by the dispersion of review times and the agreement of findings among the three reviewers who assess the same application under the same condition. Additional outcomes assess heterogeneity by reviewer, permit type, municipality, application size and review order, and are contrasted with a closing survey of reviewers. Together, these measures show whether time savings come at the expense of quality and whether AI makes reviews more uniform across professionals. All secondary outcomes are constructed from time logs, observation reports and their blind assessment, recorded identically in both arms.
|