Community-Led AI Integration for Wildfire Risk Assessment: A Participatory AI Literacy and Explainability Integration (PALEI) Framework in Los Angeles, CA

Hamed Tabkhi; Mohammad Pourhomayoun; Mona Azarbayjani; Sanaz Sadat Hosseini

arxiv: 2604.17755 · v1 · submitted 2026-04-20 · 💻 cs.CY · cs.AI

Community-Led AI Integration for Wildfire Risk Assessment: A Participatory AI Literacy and Explainability Integration (PALEI) Framework in Los Angeles, CA

Sanaz Sadat Hosseini , Mona Azarbayjani , Mohammad Pourhomayoun , Hamed Tabkhi This is my paper

Pith reviewed 2026-05-10 04:21 UTC · model grok-4.3

classification 💻 cs.CY cs.AI

keywords participatory AIwildfire risk assessmentAI explainabilitycommunity engagementmobile applicationAI literacyclimate hazardsLos Angeles

0 comments

The pith

The PALEI framework shows that early community participation and literacy building produce trusted AI tools for wildfire risk assessment.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

This paper introduces the Participatory AI Literacy and Explainability Integration (PALEI) framework to improve how AI communicates wildfire risks in urban areas like Los Angeles. Traditional tools often lose public trust because their outputs feel opaque or disconnected from local realities. The authors engaged residents in neighborhoods such as Pacific Palisades, Pasadena, and Altadena to co-design a mobile app that lets users scan visible property features for clear risk scores and tailored advice. Early results indicate strong acceptance when explanations are accessible and context-specific, along with concerns over privacy that must be addressed openly. A sympathetic reader would care because the approach offers a practical sequence for making AI useful in safety decisions by putting user understanding first rather than assuming technical accuracy alone suffices.

Core claim

The PALEI framework establishes that prioritizing early literacy building, value alignment, and participatory evaluation before model deployment leads to AI applications that achieve strong user acceptance, positive fairness perceptions, and clear adoption interest for wildfire risk communication, as demonstrated by the co-creation of a mobile application enabling residents to scan property features for interpretable risk scores with neighborhood-specific mitigation recommendations.

What carries the argument

The Participatory AI Literacy and Explainability Integration (PALEI) framework, which structures AI integration around early literacy building, value alignment, and participatory evaluation to ensure clarity, accessibility, and mutual learning between developers and residents.

Load-bearing premise

Short-term positive feedback from community workshops will translate into sustained trust, adoption, and improved preparedness once the app is deployed at scale.

What would settle it

A follow-up study measuring actual long-term app usage rates, changes in household preparedness actions, and comparisons of risk awareness between users and non-users in the same neighborhoods.

read the original abstract

Climate-driven wildfires are intensifying, particularly in urban regions such as Southern California. Yet, traditional fire risk communication tools often fail to gain public trust due to inaccessible design, non-transparent outputs, and limited contextual relevance. These challenges are especially critical in high-risk communities, where trust depends on how clearly and locally information is presented. Neighborhoods such as Pacific Palisades, Pasadena, and Altadena in Los Angeles exemplify these conditions. This study introduces a community-led approach for integrating AI into wildfire risk assessment using the Participatory AI Literacy and Explainability Integration (PALEI) framework. PALEI emphasizes early literacy building, value alignment, and participatory evaluation before deploying predictive models, prioritizing clarity, accessibility, and mutual learning between developers and residents. Early engagement findings show strong acceptance of visual, context-specific risk communication, positive fairness perceptions, and clear adoption interest, alongside privacy and data security concerns that influence trust. Participants emphasized localized imagery, accessible explanations, neighborhood-specific mitigation guidance, and transparent communication of uncertainty. The outcome is a mobile application co-designed with users and stakeholders, enabling residents to scan visible property features and receive interpretable fire risk scores with tailored recommendations. By embedding local context into design, the tool becomes an everyday resource for risk awareness and preparedness. This study argues that user experience is central to ethical and effective AI deployment and provides a replicable, literacy-first pathway for applying the PALEI framework to climate-related hazards.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

PALEI applies participatory design to LA wildfire AI but lacks solid evidence for long-term impact.

read the letter

The main point to take away is that the paper presents the PALEI framework for getting communities involved in AI wildfire risk tools in Los Angeles, but the evidence is limited to early positive feedback without enough details to back up the bigger claims about effectiveness. The new element here is applying participatory design principles specifically to wildfire risk assessment in urban Southern California neighborhoods. They describe building a mobile app where residents can get interpretable risk scores based on local features, with tailored mitigation tips. This integrates AI literacy and explainability from the start, which is a practical step for this context even if the core ideas are not brand new. It does well in identifying why traditional tools fall short on trust and accessibility. The focus on visual communication, uncertainty transparency, and community input aligns with what makes sense for real-world adoption in places like Pasadena and Altadena. Highlighting privacy concerns from participants is also a realistic touch. The soft spots are more significant. The abstract talks about strong acceptance and adoption interest but provides no participant counts, collection methods, or analysis details. This makes it difficult to assess the strength of those findings. More importantly, the paper assumes that this early engagement will translate into sustained use and better preparedness outcomes at scale. There is no longitudinal data, no before-and-after measures, and no comparison to existing risk communication methods. That assumption is the load-bearing part, and it remains untested based on what's described. The stress-test note gets this right. This paper would appeal to readers interested in participatory approaches to AI for climate hazards or disaster preparedness. Someone designing similar tools could find the process description useful as a starting point. However, it may not satisfy those looking for rigorous empirical validation or novel theoretical contributions. The work shows honest engagement with the challenges of AI deployment in sensitive areas. It deserves a serious referee to evaluate the full methods and suggest improvements, particularly around measuring real impacts. I would recommend sending it for peer review rather than desk rejecting it.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces the Participatory AI Literacy and Explainability Integration (PALEI) framework as a community-led approach to integrating AI into wildfire risk assessment for Los Angeles neighborhoods such as Pacific Palisades, Pasadena, and Altadena. It emphasizes early literacy building, value alignment, and participatory evaluation prior to model deployment, culminating in a co-designed mobile app that allows residents to scan property features for interpretable fire risk scores and tailored mitigation recommendations. The paper reports qualitative early engagement findings of strong acceptance for visual, context-specific risk communication, positive fairness perceptions, adoption interest, and concerns over privacy and data security, arguing that user experience is central to ethical AI deployment and that PALEI offers a replicable, literacy-first pathway for climate-related hazards.

Significance. If the early engagement findings were supported by detailed participant data, methods, and outcome measures, and if the framework were shown to produce measurable long-term gains in trust and preparedness, the work could offer a practical model for participatory design of AI tools in high-stakes environmental communication. The emphasis on localized imagery, accessible explanations, and transparent uncertainty communication addresses documented barriers to public trust in risk tools. However, the absence of quantitative results and validation data limits assessment of whether the approach delivers on its claims of effectiveness and replicability.

major comments (2)

[Abstract and early engagement findings] Abstract and early engagement findings section: the claims of 'strong acceptance of visual, context-specific risk communication, positive fairness perceptions, and clear adoption interest' are presented without any participant numbers, sampling method, data collection instruments, or analysis procedures. This absence leaves the central claim that PALEI supplies an effective, replicable pathway without visible empirical grounding.
[Discussion of the mobile application and outcomes] Discussion of the mobile application and outcomes: the manuscript asserts that embedding local context makes the tool 'an everyday resource for risk awareness and preparedness' but provides no pre/post measures of preparedness actions, longitudinal usage data, or comparison against existing risk communication tools to support translation from initial feedback to sustained behavioral or risk-reduction outcomes.

minor comments (2)

[Abstract] The PALEI acronym is introduced without an explicit expansion on first use in the abstract, which may reduce immediate clarity for readers unfamiliar with the framework.
[Methods] The manuscript would benefit from a dedicated methods subsection detailing how community input was elicited and incorporated, even if the current focus is descriptive.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed review of our manuscript on the PALEI framework. We address each major comment point by point below, indicating where revisions will be made to strengthen the paper while remaining faithful to the scope of this early-stage work.

read point-by-point responses

Referee: [Abstract and early engagement findings] Abstract and early engagement findings section: the claims of 'strong acceptance of visual, context-specific risk communication, positive fairness perceptions, and clear adoption interest' are presented without any participant numbers, sampling method, data collection instruments, or analysis procedures. This absence leaves the central claim that PALEI supplies an effective, replicable pathway without visible empirical grounding.

Authors: We acknowledge the need for greater transparency in reporting the early engagement methods. The findings stem from qualitative participatory workshops and interviews with residents in Pacific Palisades, Pasadena, and Altadena. In the revised manuscript we will add explicit details on participant numbers, the purposive community-based sampling approach, data collection instruments (semi-structured interviews and focus groups), and the thematic analysis procedures employed. These additions will provide the requested empirical grounding. We emphasize that the claims concern qualitative themes of acceptance and interest observed in this initial co-design phase, rather than quantitative effectiveness or full replicability validation; the framework is presented as a process model supported by these early insights. revision: yes
Referee: [Discussion of the mobile application and outcomes] Discussion of the mobile application and outcomes: the manuscript asserts that embedding local context makes the tool 'an everyday resource for risk awareness and preparedness' but provides no pre/post measures of preparedness actions, longitudinal usage data, or comparison against existing risk communication tools to support translation from initial feedback to sustained behavioral or risk-reduction outcomes.

Authors: We agree that the current phrasing risks overstating translation from initial feedback to sustained outcomes. The manuscript reports on the framework development and co-design of the app prototype; the assertion is based on participant expressions of perceived value and adoption interest during engagement sessions. We will revise the discussion section to qualify this language, clarifying that the tool is positioned as a potential everyday resource pending deployment and evaluation, and that no pre/post or longitudinal behavioral data are available at this stage. We will also note plans for future comparative studies. This addresses the concern without altering the paper's focus on the participatory process. revision: partial

Circularity Check

0 steps flagged

No significant circularity detected

full rationale

The manuscript introduces the PALEI framework as a descriptive, community-derived process for participatory AI integration in wildfire risk assessment. It contains no equations, fitted parameters, predictions, or mathematical derivations. The framework and mobile app are presented as outcomes of direct community input and early engagement findings rather than any self-referential reduction. No self-citations, uniqueness theorems, or ansatzes are invoked to support load-bearing claims. The central argument rests on reported qualitative feedback (acceptance of visuals, fairness perceptions, adoption interest) without reducing to its own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 1 invented entities

The central claim rests on the premise that participatory processes improve AI trust and usability for climate risks, which is treated as a domain assumption without independent verification in the abstract.

axioms (1)

domain assumption Community participation in AI design increases trust, acceptance, and relevance of the resulting tools
The framework and reported positive findings depend on this premise being true.

invented entities (1)

PALEI framework no independent evidence
purpose: To structure community-led AI integration for wildfire risk assessment with emphasis on literacy and explainability
The framework is introduced in this paper as a novel structured approach without prior external validation or independent evidence of its effectiveness.

pith-pipeline@v0.9.0 · 5590 in / 1389 out tokens · 71856 ms · 2026-05-10T04:21:40.064442+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

5 extracted references · 5 canonical work pages

[1]

Trends and Trajectories for Explainable, Accountable, and Intelligible Systems

“Trends and Trajectories for Explainable, Accountable, and Intelligible Systems.” In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems, 1–14. New York: ACM. Arya, V., R. K. E. Bellamy, P. Y. Chen, A. Dhurandhar, M. Hind, S. C. Hoffman, … and Y. Zhang

work page 2020
[2]

Analyzing the Risk Factors of Residential Fires in Urban and Rural Census Tracts of Ohio Using Panel Data Analysis

“Analyzing the Risk Factors of Residential Fires in Urban and Rural Census Tracts of Ohio Using Panel Data Analysis.” Applied Geography 151: 102863. Huang, X., X. Wu, and A. Usmani. 2022.“Perspectives of Using Artificial Intelligence in Building Fire Safety.” In Handbook of Cognitive and Autonomous Systems for Fire Resilient Infrastructures, 139–159. Cham...

work page 2022
[3]

Modeling Residential Development in California from 2000 to 2050: Integrating Wildfire Risk, Wildland and Agricultural Encroachment

“Modeling Residential Development in California from 2000 to 2050: Integrating Wildfire Risk, Wildland and Agricultural Encroachment.” Land Use Policy 41: 438–452. Mtani, I. W., and E. C. Mbuya

work page 2000
[4]

Accessed February 26, 2025

UL Fire Safety Research Institute. Accessed February 26, 2025 . https://fsri.org/research-update/journal-article-reports-improvements-urban-resilience-wildfires. Weidinger, L., T. Everitt, J. Rae, M. Buchanan, I. Gabriel, and others

work page 2025
[5]

Ethical and Social Risks of Harm from Language Models

“Ethical and Social Risks of Harm from Language Models.” Proceedings of the NeurIPS 2022 Workshop on ML Safety. Wexler, J., M. Pushkarna, T. Bolukbasi, M. Wattenberg, F. B. Viégas, and J. Wilson

work page 2022

[1] [1]

Trends and Trajectories for Explainable, Accountable, and Intelligible Systems

“Trends and Trajectories for Explainable, Accountable, and Intelligible Systems.” In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems, 1–14. New York: ACM. Arya, V., R. K. E. Bellamy, P. Y. Chen, A. Dhurandhar, M. Hind, S. C. Hoffman, … and Y. Zhang

work page 2020

[2] [2]

Analyzing the Risk Factors of Residential Fires in Urban and Rural Census Tracts of Ohio Using Panel Data Analysis

“Analyzing the Risk Factors of Residential Fires in Urban and Rural Census Tracts of Ohio Using Panel Data Analysis.” Applied Geography 151: 102863. Huang, X., X. Wu, and A. Usmani. 2022.“Perspectives of Using Artificial Intelligence in Building Fire Safety.” In Handbook of Cognitive and Autonomous Systems for Fire Resilient Infrastructures, 139–159. Cham...

work page 2022

[3] [3]

Modeling Residential Development in California from 2000 to 2050: Integrating Wildfire Risk, Wildland and Agricultural Encroachment

“Modeling Residential Development in California from 2000 to 2050: Integrating Wildfire Risk, Wildland and Agricultural Encroachment.” Land Use Policy 41: 438–452. Mtani, I. W., and E. C. Mbuya

work page 2000

[4] [4]

Accessed February 26, 2025

UL Fire Safety Research Institute. Accessed February 26, 2025 . https://fsri.org/research-update/journal-article-reports-improvements-urban-resilience-wildfires. Weidinger, L., T. Everitt, J. Rae, M. Buchanan, I. Gabriel, and others

work page 2025

[5] [5]

Ethical and Social Risks of Harm from Language Models

“Ethical and Social Risks of Harm from Language Models.” Proceedings of the NeurIPS 2022 Workshop on ML Safety. Wexler, J., M. Pushkarna, T. Bolukbasi, M. Wattenberg, F. B. Viégas, and J. Wilson

work page 2022