AI Climate Data Fails Without Spatial Narrative, Nanhee Kim Says

Panel 4 - 4: From AI-Generated Environmental Intelligence to Human-Centered Spatial Storytelling — Ai4

Ai4

Advanced AI environmental detection remains trapped in unread technical reports unless translated into localized spatial narratives that residents and policymakers can understand and act upon.

AI-assisted summary of Panel 4 - 4: From AI-Generated Environmental Intelligence to Human-Centered Spatial Storytelling — original source. Check the original for full context.

Source published: 2026-08-18

Detailed notes

## Background Published on YouTube by the Ai4 channel on August 18, 2026, this presentation features Nanhee Kim, an Associate Professor at California State University, Chico. According to the page description, the talk examines why AI-generated environmental intelligence rarely translates into public action and introduces a five-stage Human-Centered Spatial Storytelling Framework designed to turn technical AI analysis into engaging narrative and spatial experiences.

## TL;DR Nanhee Kim highlights the persistent disconnect between advanced AI environmental detection—such as urban heat island mapping—and the daily realities of vulnerable residents who remain unaware of data trapped in unread academic dashboards and PDF reports. Drawing on high-resolution thermal mapping in Skopje and a relatable resident persona in Phoenix, she demonstrates why abstract temperature contours fail to prompt meaningful community change on their own. She presents the Spatial Narrative Transformation Framework (SNTF), a five-stage pipeline that bridges the gap between data collection and public engagement by translating raw sensor metrics into recognizable, lived landmarks like unshaded school bus stops. Ultimately, Kim urges AI practitioners to collaborate with communication designers early, design for human comprehension, and evaluate AI initiatives by measurable community engagement rather than model accuracy alone.

## Atmosphere & style The presentation carries an urgent, humanistic, and pedagogically clear tone, effectively juxtaposing cold satellite sensor grids against evocative imagery of individuals shielding themselves from scorching heat with everyday umbrellas. Kim delivers her message with measured conviction and design-centric optimism, systematically dismantling the technical data silo into relatable human narratives. Watching the session feels both grounding and inspiring, taking an abstract technological debate and reframing it as a pressing, tangible challenge of empathy and community agency.

## Key takeaways - [00:04](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=4s) AI models can pinpoint environmental hazards down to the street level, yet those insights remain trapped in academic dashboards and research papers that the public rarely accesses. - [00:04](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=4s) The primary bottleneck in environmental communication is not the lack of AI capability, but the absence of translation between technical outputs and everyday human understanding. - [03:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=199s) Gordana Kaplan's high-resolution thermal mapping of Skopje highlights the massive disconnect between what cities look like in normal daylight and the hidden reality of urban heat accumulation. - [04:59](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=299s) The Spatial Narrative Transformation Framework (SNTF) replaces dead-end technical PDF reports with a five-stage pipeline linking AI detection to spatial narratives and public engagement. - [04:59](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=299s) While AI detects what is happening in an environment, translation through design is necessary to explain why it matters to residents' daily lives. - [07:28](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=448s) Abstract metrics like surface temperatures and tree canopy percentages become actionable when mapped to recognizable community landmarks, such as unshaded school bus stops. - [09:41](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=581s) Translating environmental data into human-centered stories establishes an impact chain where transparency builds trust, trust encourages engagement, and engagement drives informed action. - [09:41](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=581s) Organizations must move beyond static dashboards and design AI outputs centered around human relevance and narrative context. - [09:41](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=581s) Designers and communication specialists need to be integrated early into AI development rather than brought in as decorative post-processing. - [09:41](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=581s) Project success should be evaluated by community engagement and real-world utility rather than solely by model accuracy. - [12:07](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=727s) Future SNTF milestones include building interactive prototypes and conducting empirical user studies to assess public engagement in municipal sustainability initiatives. - [12:07](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=727s) Technological sophistication alone cannot solve environmental crises; lasting change requires creating human understanding and empowering people to act.

## Detailed walkthrough ### [00:04](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=4s) The Disconnect Between AI Knowledge and Lived Experience

Nanhee Kim opens by presenting an aerial thermal map of an urban area and asking the audience to identify the hottest neighborhood. While modern artificial intelligence already knows the exact answer down to the street level, the residents living on those very blocks remain completely unaware. Environmental intelligence currently remains trapped inside datasets, academic research papers, and complex analytical dashboards that "nobody outside the lab opens." This creates a profound chasm between what AI models detect and what human beings physically experience on the ground.

```mermaid flowchart LR accTitle: The Current Environmental Data Dead End vs Human Reality subgraph DataSilo["AI Data Silo"] AI["AI Model & Satellites"] --> Repo["Academic Papers & Dashboards"] Repo --> DeadEnd["Dead End: Unopened File"] end subgraph StreetReality["Lived Experience"] Person["Resident on the Hottest Block"] --> Tool["Makeshift Shade: Umbrella"] end ```

To illustrate the human cost of this divide, Kim introduces the image of a resident walking to work under the scorching sun, holding an umbrella for shade because "it is the only tool they have." Meanwhile, just miles away, advanced AI models hold detailed, actionable solutions stored away in unread files. The central issue in environmental communication is not the need for more sophisticated AI models—AI is already exceptionally capable of detecting urban heat islands, vegetation loss, and water quality degradation—but rather the critical lack of translation between technical outputs and everyday human understanding.

- **Speaker Affiliation**: Nanhee Kim, Associate Professor in the Department of Media Arts and Design Technology at California State University, Chico. - **Core Premise**: AI provides accurate environmental data, but without human translation, residents cannot understand or act upon those insights.

[03:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=199s) Two Views of the Same City: The Gordana Kaplan Case Study

To ground the visual reality of this information gap, Kim highlights an aerial dual-image of Skopje. The visual juxtaposes a standard aerial photograph of the city with a high-resolution thermal imaging map captured simultaneously. This was not a generic stock asset, but the work of PhD researcher Gordana Kaplan, who flew over her own hometown in August 2018 aboard a small Cessna equipped with thermal cameras.

Kaplan stitched together 20,000 individual aerial images to construct a seamless 1.5-meter resolution surface temperature map. While the visible-light view looks like a normal summer day, the thermal layer exposes a dramatically different, intense reality of heat accumulation across the urban fabric. This stark visual comparison represents the overarching challenge: technical detection sits on one end, public understanding sits on the other, and there is currently no established intermediate bridge connecting the two.

- **Location & Date**: Skopje, North Macedonia; August 2018. - **Methodology**: 20,000 thermal aerial images captured from a Cessna aircraft. - **Image Resolution**: 1.5-meter real surface temperature dataset.

[04:59](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=299s) From Dead-End PDFs to the Spatial Storytelling Framework

Under the standard paradigm, environmental data follows a linear path from satellite sensors and AI models directly into technical PDF reports—a functional dead end. Kim introduces the Human-Centered Spatial Storytelling Framework, also referred to as the Spatial Narrative Transformation Framework (SNTF), to redirect that pipeline toward human comprehension and community action.

```mermaid flowchart TD accTitle: Human-Centered Spatial Storytelling Pipeline subgraph Stage1["1. Detection"] A["Satellite & Drone Sensing"] --> B["AI Analysis & Heat Mapping"] end subgraph Stage2["2. Spatial Narrative Transformation"] B --> C["Translation: Contextualized Meaning"] C --> D["Spatial Storytelling & Experience"] end subgraph Stage3["3. Public Impact"] D --> E["Public Engagement"] E --> F["Community Action & Policy Decisions"] end ```

The five-stage framework takes environmental data through an intentional sequence of satellite scanning, AI analysis, narrative structuring, spatial experience, and public engagement. Kim emphasizes that "AI only tells us what's happening," but raw detection does not explain *why* it matters to a resident's daily life. The framework relies on translation as an active design discipline rather than an afterthought, ensuring that data culminates in public conversation rather than an archival report.

- **Current Pipeline**: Satellite $\rightarrow$ AI $\rightarrow$ PDF Report $\rightarrow$ Dead End. - **SNTF Pipeline**: Satellite/Drones $\rightarrow$ AI Insight $\rightarrow$ Spatial Narrative $\rightarrow$ Human Experience $\rightarrow$ Action.

[07:28](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=448s) Translating Data into Human Meaning: Sarah's Journey in Phoenix

Kim demonstrates how translation operates through a concrete persona: Sarah, a resident of Phoenix who simply wants her neighborhood to be cooler. Sarah does not read technical AI whitepapers, nor does she need more raw data. When an aerial drone scans her neighborhood and flags her street as a severe heat island, the technical heatmap alone offers her no actionable context.

```mermaid flowchart LR accTitle: Data vs Meaning Translation Example subgraph RawData["Raw AI Intelligence"] R1["Surface Temperature Data"] R2["Tree Canopy Percentage"] R3["Abstract Heat Contours"] end subgraph HumanMeaning["Human-Centered Spatial Story"] M1["School Bus Stop Lacks Shade"] M2["Children Exposed to Danger"] M3["Targeted Tree Planting Site"] end RawData -->|Spatial Narrative Translation| HumanMeaning ```

Through spatial storytelling, abstract metrics like surface temperature contours and tree canopy percentages are translated into recognizable physical landmarks. Sarah sees the street corner where children wait for the school bus in unshaded, dangerous heat, recognizing that planting shade trees at that specific spot makes the route safe. The underlying dataset remains identical, but contextualizing it transforms raw heat figures into shared community purpose.

- **Persona**: Sarah, living in Phoenix, navigating extreme urban heat. - **Narrative Shift**: Moving from abstract temperature gradients to identifying vulnerable sites such as unshaded school bus stops.

[09:41](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=581s) The Chain of Impact and Four Practical Takeaways

Translating intelligence into narrative produces a direct behavioral chain: transparency builds trust, trust encourages public engagement, and engagement drives informed decision-making. Kim notes that communication design and AI development have historically operated in silos, barely speaking to one another. SNTF serves as an operational bridge so interdisciplinary teams can implement Human-Centered AI in practice.

```mermaid flowchart LR accTitle: The Trust to Action Chain A["Transparency"] --> B["Trust"] B --> C["Engagement"] C --> D["Better Decisions & Action"] ```

Kim outlines four practical steps that organizations and AI practitioners can adopt immediately: - **Go beyond the dashboard**: Move past static charts and tabular visualizations that only technical analysts can decipher. - **Design for understanding**: Center outputs around narrative context and user relevance rather than pure information density. - **Involve designers early in the AI pipeline**: Treat communication and environmental graphic design as foundational translation tools rather than decorative post-processing. - **Measure engagement, not just model accuracy**: Evaluate AI projects by whether target communities understand and utilize the findings to effect change.

[12:07](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=727s) Future Roadmap and the Core Philosophy of SNTF

Looking toward future work, Kim explains that while the Spatial Narrative Transformation Framework is currently an interdisciplinary conceptual model, the next phase focuses on building interactive prototypes and conducting empirical research. Following a recently published journal paper, Kim plans to run user studies to evaluate how effectively spatial narratives help diverse communities understand, engage with, and act on environmental intelligence in real-world sustainability projects.

Kim concludes with a reminder that technological sophistication alone cannot solve environmental crises. AI can detect patterns with unprecedented speed and precision, but "knowledge alone doesn't change cities" or alter human behavior—people do. The value of environmental intelligence is realized only when individuals understand the data, care about the outcome, and possess the agency to act.

- **Upcoming Milestones**: Interactive prototype development, empirical user testing, and field deployment in municipal sustainability initiatives. - **Concluding Philosophy**: "The future of AI isn't only about building better intelligence. It's about creating better understanding."

## Notable quotes > [00:04](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=4s) nobody outside the lab opens — Nanhee Kim

> [00:04](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=4s) it is the only tool they have. — Nanhee Kim

> [04:59](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=299s) AI only tells us what's happening, — Nanhee Kim

> [09:41](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=581s) Go beyond the dashboard — Nanhee Kim

> [09:41](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=581s) Design for understanding — Nanhee Kim

> [09:41](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=581s) Involve designers early in the AI pipeline — Nanhee Kim

> [09:41](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=581s) Measure engagement, not just model accuracy — Nanhee Kim

> [12:07](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=727s) knowledge alone doesn't change cities — Nanhee Kim

> [12:07](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=727s) The future of AI isn't only about building better intelligence. It's about creating better understanding. — Nanhee Kim

## Mentioned Nanhee Kim, California State University, Chico, Department of Media Arts and Design Technology, Ai4, Gordana Kaplan, Cessna, Skopje, North Macedonia, Phoenix, Human-Centered Spatial Storytelling Framework (Spatial Narrative Transformation Framework / SNTF)

## Video description Nanhee Kim, Associate Professor, California State University, Chico

AI can now detect a warming neighborhood or a stressed watershed before anyone sets foot in it—but why do those insights so rarely become experiences that people understand and act upon? In this talk, I'll explore why AI-generated environmental intelligence often fails to reach the public in meaningful and engaging ways. Drawing on Human-Centered AI and environmental graphic design, I'll present a five-stage Human-Centered Spatial Storytelling Framework that transforms AI analysis into narrative, spatial experience, and public engagement, illustrated through an urban heat island case study. Attendees will leave with a practical model for translating AI outputs into experiences that people can understand, trust, and act on. Whether you work in AI, design, sustainability, or environmental communication, you'll gain a framework for making AI insights more accessible and impactful.

[Source](https://www.youtube.com/watch?v=S_mbNTfnS-U)

中文简报

Nanhee Kim：若无空间叙事，AI气候数据难见实效先进的AI环境检测若不转化为居民和决策者能理解并行之有效的本地化空间叙事，就仍将被困在无人问津的技术报告中。 Nanhee Kim指出，先进的AI环境检测（如城市热岛制图）与脆弱群体的日常生活现实之间长期存在脱节，而后者对困在无人阅读的学术仪表板和PDF报告中的数据一无所知。结合Skopje的高分辨率热成像制图以及Phoenix贴近生活的居民画像，她论证了为何仅凭抽象的等温线无法促成有意义的社区变革。为此，她提出了空间叙事转化框架（SNTF），该框架通过包含五个阶段的流程，将原始传感器指标转化为无遮阴校车站等易于辨识的生活地标，从而架起数据采集与公众参与之间的桥梁。最后，Kim敦促AI从业者及早与传播设计师合作，面向人类理解力进行设计，并以可衡量的社区参与度而不仅凭模型准确率来评估AI项目的成效。 AI 模型已能将环境风险精准定位到街道级别，但这些洞察大多仍被困在极少有公众访问的学术论文与专业仪表盘中。 (0:04) 研究员 Gordana Kaplan 对斯科普里（Skopje）的高分辨率热成像图显示，常规日照下的城市街景与深层城市热岛聚集的现实之间存在着巨大断层。 (3:19) 空间叙事转化框架（SNTF）旨在打破技术 PDF 报告的沟通死胡同，通过五阶段流程将 AI 环境检测与空间叙事及公众参与相串联。 (4:59) 地表温度与树冠覆盖率等抽象指标在被映射到无遮阴校车站等具体社区地标后，才能转化为居民能够理解并据此采取行动的有效信息。 (7:28) 将环境数据转化为以人为本的故事能够建立完整的影响力链条：透明度构建信任，信任促进公众参与，而参与则推动知情行动与决策。 (9:41)

Reading summary

Associate Professor Nanhee Kim argues that environmental AI's real-world impact is constrained not by technical detection limits, but by a failure to translate complex datasets into accessible spatial narratives that everyday communities can understand and act upon. The Disconnect Between AI Detection and Street Reality (0:04) Modern AI models and remote sensing systems can detect urban heat islands, tree canopy loss, and water quality issues down to the street level, yet these insights remain largely confined to academic papers and analytical dashboards that outside audiences never open. Kim illustrates this divide with the image of a resident walking to work under the sun with an umbrella for makeshift shade, completely unaware that nearby AI models have already mapped their exact block as an extreme heat zone. To demonstrate this communication gap, Kim points to a case study by PhD researcher Gordana Kaplan, who flew over Skopje, North Macedonia, in August 2018 aboard a Cessna equipped with thermal cameras. Kaplan stitched 20,000 aerial images into a 1.5-meter resolution surface temperature map. While the visible-light view showed an ordinary summer day, the thermal layer revealed acute heat concentration across the urban fabric, exemplifying the lack of an intermediate bridge between technical detection and public awareness. The Spatial Narrative Transformation Framework (4:59) Under the current environmental data model, findings flow from satellites and AI models directly into technical PDF reports—a path Kim characterizes as a functional dead end. In response, Kim introduces the Spatial Narrative Transformation Framework (SNTF), also termed the Human-Centered Spatial Storytelling Framework, to redirect this sequence across five stages: sensing via satellites and drones, AI analysis, narrative structuring, spatial experience, and public engagement. Kim emphasizes that while AI demonstrates what is occurring, it does not explain why those findings matter to a resident's daily life. SNTF approaches translation as an active design discipline rather than an afterthought, aiming to convert technical outputs into public conversations, community action, and informed municipal policy decisions. Translating Abstract Metrics into Recognizable Landmarks (7:28) To show how narrative translation works in practice, Kim examines the persona of Sarah, a Phoenix resident coping with extreme urban heat. Sarah does not read academic whitepapers and cannot derive practical steps from abstract heat contours or canopy percentages alone. Spatial storytelling translates those remote metrics into physical landmarks. Rather than seeing raw temperature gradients, Sarah sees a specific unshaded school bus stop where children wait in dangerous heat, identifying an immediate location for targeted tree planting. While the underlying dataset remains unchanged, contextualizing the information transforms raw metrics into a tangible, shared community purpose. Four Practical Imperatives for AI Practitioners (9:41) Kim notes that communication design and AI development have historically operated in isolated silos. To connect them, Kim outlines four practical directives for organizations: move beyond static tabular dashboards, design outputs around narrative context and user relevance, integrate communication and environmental graphic designers early in the AI workflow, and measure project outcomes by community engagement rather than model accuracy alone. While the Spatial Narrative Transformation Framework is currently an interdisciplinary conceptual model, Kim plans to build interactive prototypes and conduct empirical user studies in municipal sustainability projects following a recent journal paper. Kim concludes that technical precision alone cannot solve environmental crises, as lasting urban change requires that residents understand the data, care about the problem, and have the agency to respond.

精炼阅读摘要

加州州立大学奇科分校的Nanhee Kim副教授指出，现代人工智能虽已具备街道级的环境监测能力，但技术成果多受限于难以理解的报告与仪表板中；唯有建立以人为本的空间叙事转译机制，将抽象数据转化为具体的生活场景体验，才能弥合技术与公众行动之间的鸿沟。环境智能与公众感知的脱节现实 (0:04) 根据Kim的阐述，当前环境智能领域面临的核心矛盾并非AI模型能力的不足，而是先进数据洞见与居民日常体感之间的深刻鸿沟。现代卫星传感器和AI算法已能精准识别城市热岛效应、植被流失和水质退化，甚至精确锁定城市中最炎热的街区。然而，生活在这些高温街区的居民对此一无所知，在缺乏有效信息的情况下，往往只能依赖撑雨伞遮阳等原始手段应对酷热。Kim指出，包含切实解决方案的高精度环境数据目前大多被困在学术论文、研究报告和复杂的数据分析仪表板内，除了实验室的研究人员外几乎无人查阅，最终在未被打开的PDF文件中走向终点。为了直观证明这种信息脱节，Kim展示了博士研究员Gordana Kaplan于2018年8月在北马其顿斯科普里（Skopje）开展的案例研究。Kaplan搭乘一架装备了热成像相机的塞斯纳（Cessna）小型飞机飞越家乡上空，拼接了2万张航空影像，绘制出一张分辨率达1.5米的高精度真实地表温度图。普通航拍照呈现的是看似寻常的夏日景象，而热成像图则揭示出建筑物与街道极端积热的严峻现实。Kim强调，这一对比凸显了当前技术端与公众认知端的分离：技术检测已具备极高的精度，但中间缺乏成熟的转化机制，导致科学洞见无法服务于受影响的人群。空间叙事转化框架重塑数据流转 (4:59) 针对环境数据流向封闭技术报告的死循环，Kim提出了以人为本的空间叙事转化框架（Spatial Narrative Transformation Framework，简称SNTF，亦称空间叙事框架）。在传统的线性工作流中，数据从卫星传感器与AI模型直接进入技术性PDF报告，缺乏触达基层的有效路径。SNTF重构了这一链路，将其划分为涵盖卫星与无人机探测、AI分析、叙事结构构建、空间体验传达以及公众参与的五个阶段，旨在将环境数据由静态归档引向具体的社区行动与政策制定。 Kim在此框架中强调，“AI只能告诉我们发生了什么”，但原始的数据检测并不能自动解释这些现象为何对居民的切身利益至关重要。因此，SNTF主张将信息转译视为一门主动的设计学科，而非技术研发后的装饰性修饰。通过在技术分析与最终应用之间构建叙事纽带，该框架确保环境智能最终能够转化为公众讨论与治理举措，改变技术成果仅仅停留在实验室内部的现状。将抽象指标转化为具象社区意义 (7:28) 为展示空间叙事在实践中的转译机制，Kim构建了生活在亚利桑那州凤凰城的虚拟居民Sarah这一案例。面对极端城市热浪，Sarah希望改善街区微气候，但她既不会阅读复杂的AI技术白皮书，也不需要更多晦涩的原始数据。当无人机扫描街区并将她的街道标注为重度热岛时，色块斑斓的技术热力图并不能为她提供可以直接理解或付诸行动的现实背景。通过空间叙事设计，研究团队将地表温度等值线和树冠覆盖率百分比等高度抽象的指标，转译为居民日常可见的具体物理地标与生活情境。例如，系统不再仅仅展示温度梯度，而是明确指出孩子们无遮蔽等待校车的街角公交车站正处于危险高温之下，并提示在该特定位置种植遮荫树木可以显著改善路线安全。Kim指出，底层的数据源并没有发生变化，但通过将数据嵌入到具体的空间生活叙事中，原本冰冷的温度数字被成功赋予了社区共同关注的现实意义。从技术指标迈向公众参与的行动路径 (9:41) Kim提出，将智能数据转译为叙事体验能够触发一条明确的社会行为影响链：数据透明性能够建立公众信任，信任能激发基层参与，而广泛的参与则最终促成明智的决策与集体行动。然而，传播设计与AI技术研发长期处于相互割裂的状态，二者之间缺乏深度沟通。Kim认为，SNTF正是一个可操作的跨学科桥梁，能够将“以人为本的AI”理念真正落实到工作流程中。针对各类机构与AI从业人员，Kim归纳了四项可立即推行的实践原则：第一，超越静态仪表板，摆脱只有数据分析师才能读懂的图表与表格；第二，以理解为导向进行设计，聚焦输出结果的叙事背景和用户相关性，而非单方面追求信息密度；第三，在AI研发早期即引入设计师，将传播学与环境平面设计视为核心转译工具，而非项目后期的美化包装；第四，将衡量标准从纯粹的模型精度拓展至公众参与度，评估目标社区是否真正理解并运用分析结果推动了实质改变。 Kim展望表示，SNTF框架目前处于跨学科概念阶段，未来研究将专注于开发交互式原型并开展实证用户测试，以检验空间叙事在不同社区可持续发展项目中的落地成效。她总结认为，技术复杂性本身无法解决环境危机，算法能以空前速度识别模式，但改变城市与行为的决定性力量依然是人类自身；只有当公众理解数据、关心结果并具备行动能力时，环境智能的真实价值才得以实现。

Detailed notes — zh

## Background 该视频由 Ai4 频道于 2026 年 8 月 18 日发布在 YouTube 平台。内容记录了加利福尼亚州立大学奇科分校副教授金南熙（Nanhee Kim）的主题演讲，探讨 AI 环境洞察为何难以触达公众，并系统介绍了如何通过五阶段“以人为本的空间叙事框架”将 AI 分析转化为大众可理解、可信任且能付诸实践的空间体验与行动。 ## TL;DR 尽管尖端 AI 能够以极高精度检测城市热岛等环境危机，但技术洞察往往受困于专业报告与仪表盘中，无法激发公众共鸣与现实行动。金南熙提出了“以人为本的空间叙事框架”（SNTF），核心在于将冰冷的地表温度与卫星数据重构为关乎切身生活的人类情境叙事。通过菲尼克斯居民 Sarah 的案例演练，演讲阐明了从直观认知、建立信任到驱动社区参与决策的递进闭环。最后，她向跨学科从业者提出建议，呼吁设计师及早介入 AI 研发，并强调应以公众参与度和理解度而非单纯的模型精度来衡量 AI 的社会价值。 ## Atmosphere & style 整场演讲兼具严谨的学术洞见与温润的人文关怀，语气清晰平实且富有感染力。金南熙巧妙地利用烈日下撑伞的行人和斯科普里震撼的航拍热成像对比图，营造出强烈的认知张力与现实紧迫感。从剖析“数据死胡同”的无奈现实，到展示叙事化重塑后的社区生机，演讲在层层递进中完成了从理论批判到实践赋能的情感跃升，令观众深受启发。 ## Key takeaways - [00:04](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=4s) AI 在环境洞察方面已极其敏锐，但这些数据大多被锁在学术论文、专业数据集与仪表盘中，与饱受高温困扰的普通公众现实之间存在巨大鸿沟。 - [02:00](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=120s) 当前解决环境危机所缺失的环节并非更强大的 AI 算法，而是将冰冷技术指标转化为公众理解的“更好的转译（better translation）”。 - [03:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=199s) 戈尔达娜·卡普兰在北马其顿斯科普里的航拍热成像研究表明，顶尖 AI 环境监测与公众认知断层并存，传统数据流向极易走向无人问津的“死胡同”。 - [06:01](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=361s) 金南熙正式提出“以人为本的空间叙事框架”（SNTF），将 AI 分析、叙事构建、空间体验与公众参与融为一体，致力于从纯数据分析跃升至意义传达。 - [07:00](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=420s) SNTF 框架的核心在于“人类故事”，通过重构环境数据与日常生活的联系，让公众明白环境指标为何与切身利益息息相关。 - [08:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=499s) 以菲尼克斯居民 Sarah 为例，结合街景与生活情境的空间叙事远比抽象热力图更能触动居民，推动公众建立真正的认知。 - [09:20](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=560s) 推动社区改善遵循严密的递进逻辑链：数据透明度建立信任，信任激发公众参与，广泛参与最终催生切实的改善行动。 - [10:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=619s) 空间叙事并非科研产出后的后期视觉装饰，而是一套将“以人为本的 AI”转化为跨学科可操作范式的系统转译实践。 - [11:10](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=670s) 金南熙呼吁从业者超越传统仪表盘、为理解而设计、让设计师尽早介入 AI 研发，并以公众参与度而非单纯的模型精度作为价值衡量标准。 - [12:00](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=720s) 知识本身无法直接改变城市与社区，唯有通过转译使人们真正理解并产生共鸣，才能驱动改变发生。 ## Detailed walkthrough ### [00:04](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=4s) AI 洞察与公众现实之间的鸿沟演讲伊始，加利福尼亚州立大学奇科分校（California State University, Chico）媒体艺术与设计技术系副教授金南熙（Nanhee Kim）向观众展示了一张城市地图并抛出提问：“大家觉得这张地图上哪个街区最热？”现场观众很快察觉到了答案，而实际上，AI 早就一清二楚。然而，令人遗憾的现实是：绝大多数居民根本看不到这些数据。这些宝贵的洞察通常被深锁在数据集、学术论文以及实验室之外无人问津的仪表盘里。生活在整座城市最炎热街区的人们，压根不知道关于自己脚下街道的高温数据早已存在。AI 所掌握的客观事实与普通人实际感知到的现实之间，存在着巨大的鸿沟。金南熙进一步刻画了现实生活中的具体图景：真正面对环境危机的是一个活生生的人，而不是地图上的像素点或电子表格里的一行行数据。画面中，一位行人在烈日下艰难地走去上班，手里撑着一把遮阳伞——因为这是他们手中唯一能用来遮阴防暑的工具。与此同时，仅仅几英里之外的实验室里，某个 AI 模型早已计算出了精准得多的数据与应对方案，但这份方案却静静地躺在一个无人发送的文件里。当今的 AI 在环境问题识别上已经变得极其敏锐，无论是热岛效应、植被生长状态还是水体质量，它都能以不可思议的精细度进行分析。但当这些信息最终传达到公众面前时，往往只是晦涩的技术报告、复杂的仪表盘或专业地图。这些工具对分析师而言固然出色，但对普通大众来说却如同天书，既难以理解更无法激发行动。金南熙指出，当前缺失的环节从来不是“更好的 AI 算法”，而是“更好的转译（better translation）”。 ### [03:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=199s) 斯科普里的航拍对比：相同地点，两副面孔为了直观展现这一鸿沟，金南熙分享了一组极具冲击力的真实航拍研究图像。画面下半部分是从空中俯瞰北马其顿首都斯科普里（Skopje）的普通航拍图，上半部分则是同一座城市在盛夏酷暑中的地表温度热成像图。这项研究由博士研究员戈尔达娜·卡普兰（Gordana Kaplan）完成。2018 年 8 月，卡普兰驾驶一架小型塞斯纳（Cessna）飞机飞越自己的家乡上空，利用热成像相机拍摄并拼接了整整 20,000 张图像，最终合成了分辨率高达 1.5 米的真实地表温度分布图。这是发表在学术期刊上的严谨科研成果，绝非凭空捏造的素材图片。同一地点、同一瞬间，两张截然不同的图像赤裸裸地揭示了地表正在发生的一切。这张对比图恰恰映射出了核心困境：上方是尖端的 AI 环境监测，下方是迷茫的公众认知，而二者之间空无一物，没有任何桥梁将技术认知转化为大众理解。传统环境数据流向与金南熙提出的框架路径形成了鲜明对照： ```mermaid flowchart TD accTitle: 传统数据路径与空间叙事框架路径对比 subgraph Traditional["传统路径：数据死胡同"] A["卫星/传感器数据"] --> B["AI 分析与算法模型"] B --> C["学术报告 / PDF / 仪表盘"] C --> D["实验室死胡同（无人问津）"] end subgraph Proposed["空间叙事框架（SNTF）路径"] E["卫星/传感器数据"] --> F["AI 洞察"] F --> G["人类叙事转化"] G --> H["空间沉浸体验"] H --> I["公众理解与社区行动"] end ``` - **数据采集规格**：戈尔达娜·卡普兰团队于 2018 年 8 月通过轻型塞斯纳飞机挂载热成像仪采集，拼接了 20,000 张影像，地表温度空间分辨率达到 1.5 米。 - **现存痛点**：传统模式下环境数据止步于 PDF 报告或专业仪表盘；全新框架则旨在让数据最终进入产生共鸣与理解的公众心中，让原本的死胡同演变为广泛的公共对话。 ### [06:01](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=361s) 以人为本的空间叙事框架：从数据到意义面对如何弥合上述断层的问题，金南熙正式提出了“以人为本的空间叙事框架”（Human-Centered Spatial Storytelling Framework，简称 SNTF）。该框架并不满足于停留于 AI 生成的数据洞察，而是致力于将技术分析深度转化为大众易于理解且能付诸实践的内容。它巧妙地将 AI 分析、叙事构建、空间体验以及公众参与串联为一个有机整体。金南熙强调："AI 只负责告诉我们发生了什么，但仅仅知道发生了什么还远远不够；公众必须明白这为什么与自己的生活息息相关。" 整个转译框架由五个递进阶段构成，其中最为核心的一环正是“人类故事（human story）”。金南熙对比了转译前后的不同呈现：在左侧，AI 提供的是纯粹的技术指标——地表温度数值、树冠覆盖率以及五彩斑斓的热力图。这些指标虽极度精准，却无法触动人心，无法说明公众为何应当对此予以关注。而在右侧，核心数据本身并没有变动，发生巨变的只有其“意义（meaning）”。转译后的信息以贴近人性的方式直白呈现：“这是孩子们每天顶着烈日等待校车的地方，这里几乎没有任何树荫遮挡；如果在此处种上几棵树，整条街道的安全度和舒适度就会大幅提升。”同样的数据，经由场景化叙事与空间联系重构后，瞬间变得清晰易懂、关乎切身利益。 ### [08:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=499s) 案例演练：菲尼克斯居民 Sarah 的认知与行动闭环为了让理念更加落地，金南熙引入了一位生活在亚利桑那州菲尼克斯（Phoenix）的虚构居民 Sarah。Sarah 从不阅读冗长枯燥的 AI 气候报告，她的诉求纯粹而朴素——只希望自己生活的街区能变得凉快一些。她所需要的并不是更多的算法模型，而是对所处环境切实、清晰的理解。 Sarah 的体验旅程始于低空无人机对整个社区的高精度扫描。AI 敏锐地圈定出了温度最高的几条酷暑街道，但对 Sarah 而言，一张花花绿绿的热力图依然毫无意义。此时，空间叙事机制开始介入：系统不再向 Sarah 推送抽象图表，而是呈现她所熟悉的街景。她看到了自己日常走过的拐角、等车的车站，瞬间明白了极端高温是如何实实在在地侵蚀着社区邻里的日常生活。当客观数据被赋予了生活实感，公众的理解便在这一刻真正生根发芽。金南熙着重阐述了从认知到实践的链式反应机制：社区民众从来不会为了冰冷的地表温度分布图而行动，驱动他们采取行动的永远是赋予数据温度的故事。这种转化遵循着严密的递进逻辑： 1. **透明度（Transparency）**：清晰直观的呈现方式建立起公众对数据的信任。 2. **信任（Trust）**：信任鼓励居民积极投入社区交流与讨论。 3. **公众参与（Engagement）**：广泛的参与催生出更科学、更合理的社区决策。 4. **行动（Action）**：最终推动街道植树、遮阳设施建设等切实的改善落地。 ### [10:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=619s) 跨学科融合与行动建议：构建 SNTF 的未来金南熙回顾指出，过去很长一段时间里，AI 智能与信息传播设计（Communication Design）几乎处于各自为政的割裂状态。以人为本的空间叙事方法论并不是一种在科研产出后临时涂抹的“视觉装饰”，而是一套扎实严谨的系统转译实践。它赋予了跨学科团队一套可落地的操作范式，将看似虚无缥缈的“以人为本的 AI（Human-Centered AI）”落地为各组织机构能够切实操作的工作流程。面向现场来自 AI、设计、可持续发展及传播领域的从业者，金南熙给出了四条立即可行的落地建议： - **超越传统仪表盘（Go beyond the dashboard）**：摆脱对单一数据大屏或静态 PDF 的依赖。 - **为理解而设计（Design for understanding）**：将设计的重心从单纯的“信息呈现”转向“促进人的深层理解”。 - **让设计师尽早介入 AI 研发流程（Bring designers into the AI process early）**：打破技术与体验的壁垒，让跨学科视角贯穿模型构思全程。 - **衡量参与度而不仅是模型精度（Measure engagement, not just model accuracy）**：算法的价值绝不单纯取决于准确率指标，唯有当公众能够充分理解并据此采取行动时，AI 才能真正创造社会价值。展望未来，金南熙表示该空间叙事转化框架（Spatial Narrative Transformation Framework, SNTF）已完成了跨学科理论构建，并在相关学术期刊上发表了论文。接下来的核心工作是开发交互式原型系统，并在真实的社区可持续发展项目中开展实证分析，量化检验该方法在提升公众理解、情感共鸣与行动参与方面的实际成效。金南熙在掌声中总结道：“知识本身无法改变城市，也无法改变人类行为或社区；能改变这一切的只有人。当人们真正理解并开始在乎时，行动才会发生。” ## Notable quotes > [00:04](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=4s) 大家觉得这张地图上哪个街区最热？ — 金南熙 > [00:04](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=4s) 更好的转译 — 金南熙 > [06:01](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=361s) AI 只负责告诉我们发生了什么，但仅仅知道发生了什么还远远不够；公众必须明白这为什么与自己的生活息息相关。 — 金南熙 > [06:01](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=361s) 这是孩子们每天顶着烈日等待校车的地方，这里几乎没有任何树荫遮挡；如果在此处种上几棵树，整条街道的安全度和舒适度就会大幅提升。 — 金南熙 > [10:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=619s) 超越传统仪表盘 — 金南熙 > [10:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=619s) 为理解而设计 — 金南熙 > [10:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=619s) 让设计师尽早介入 AI 研发流程 — 金南熙 > [10:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=619s) 衡量参与度而不仅是模型精度 — 金南熙 > [10:19](https://www.youtube.com/watch?v=S_mbNTfnS-U&t=619s) 知识本身无法改变城市，也无法改变人类行为或社区；能改变这一切的只有人。当人们真正理解并开始在乎时，行动才会发生。 — 金南熙 ## Mentioned 金南熙（Nanhee Kim）, 加利福尼亚州立大学奇科分校（California State University, Chico）, 戈尔达娜·卡普兰（Gordana Kaplan）, 斯科普里（Skopje）, 塞斯纳（Cessna）, 菲尼克斯（Phoenix）, 以人为本的空间叙事框架（SNTF）, 以人为本的 AI（Human-Centered AI）, Ai4 ## Video description Nanhee Kim, Associate Professor, California State University, Chico AI can now detect a warming neighborhood or a stressed watershed before anyone sets foot in it—but why do those insights so rarely become experiences that people understand and act upon? In this talk, I'll explore why AI-generated environmental intelligence often fails to reach the public in meaningful and engaging ways. Drawing on Human-Centered AI and environmental graphic design, I'll present a five-stage Human-Centered Spatial Storytelling Framework that transforms AI analysis into narrative, spatial experience, and public engagement, illustrated through an urban heat island case study. Attendees will leave with a practical model for translating AI outputs into experiences that people can understand, trust, and act on. Whether you work in AI, design, sustainability, or environmental communication, you'll gain a framework for making AI insights more accessible and impactful. [Source](https://www.youtube.com/watch?v=S_mbNTfnS-U)

Source quotes

[0:04] 金南熙: 大家觉得这张地图上哪个街区最热？ [0:04] 金南熙: 更好的转译 [6:01] 金南熙: AI 只负责告诉我们发生了什么，但仅仅知道发生了什么还远远不够；公众必须明白这为什么与自己的生活息息相关。 [6:01] 金南熙: 这是孩子们每天顶着烈日等待校车的地方，这里几乎没有任何树荫遮挡；如果在此处种上几棵树，整条街道的安全度和舒适度就会大幅提升。 [10:19] 金南熙: 超越传统仪表盘 [10:19] 金南熙: 为理解而设计 [10:19] 金南熙: 让设计师尽早介入 AI 研发流程 [10:19] 金南熙: 衡量参与度而不仅是模型精度 [10:19] 金南熙: 知识本身无法改变城市，也无法改变人类行为或社区；能改变这一切的只有人。当人们真正理解并开始在乎时，行动才会发生。

Source quotes

[0:04] Nanhee Kim: nobody outside the lab opens [0:04] Nanhee Kim: it is the only tool they have. [4:59] Nanhee Kim: AI only tells us what's happening, [9:41] Nanhee Kim: Go beyond the dashboard [9:41] Nanhee Kim: Design for understanding [9:41] Nanhee Kim: Involve designers early in the AI pipeline [9:41] Nanhee Kim: Measure engagement, not just model accuracy [12:07] Nanhee Kim: knowledge alone doesn't change cities [12:07] Nanhee Kim: The future of AI isn't only about building better intelligence. It's about creating better understanding.

Original transcript

[0:04] So, yeah, here's my equation, right? So, which neighborhood do you think [0:08] is the hottest on this map? So, yeah, I know like some of you already [0:13] noticed and then we saw AI already knows, right? Yeah, he already has the highest, right? Mhm. [0:24] Woah. But, um mostly resident never see this. And then it stays in like a Nela set. A research paper [0:36] and dashboard nobody outside the lab opens. And then the people living the hottest block [0:43] of their own city have no idea the data about their own street already exist. So, the gap [0:52] and between what AI knows and then what people experience is exactly what this talk is about. [1:04] Awesome. So, good morning, right, everyone? So, my name is Nani Kim [1:10] and from the Department of Media Arts and Design Technology at California State University, Chico. [1:16] And let's start with the like neighborhood you just saw and AI can already tell us [1:24] what's happening there. But, if the people who live there don't understand those insights, [1:31] they can't act on them. So, that's the challenge that my research address and then today I'd like [1:40] to show you how we can bridge that gap through a human-centered special storytelling framework. [1:52] So, this is who this is. Not a pixel on a map, not a row in a spreadsheet. A person walking to work [2:06] and then using an umbrella as shade because it is the only tool they have. And then while [2:14] a few miles away, an AI model already has far better answer sitting in a file nobody sent them. [2:28] So, AI getting very good at detecting environmental problems. [2:35] It can identify heat island and then vegetation state and water quality in incredible detail. [2:46] But, when that information reaches people, it usually comes as reports and dashboards or maps. [2:57] Those are great for analysts. They are like much harder for the public to understand it and act on. [3:08] The missing place isn't better AI, it's better translation. So, that's what the research about. [3:19] So, this is the one real photo and split into the bottom is the one you would see flying over [3:28] Skopje. The top is the exact same city in summer. And I want to tell you like who made it is briefly [3:39] and because it matters. And a PhD researcher named Gordana Kaplan flew over her own hometown [3:48] in a small Cessna with a thermal cameras in August like 2018 and stitched together 20,000 [3:58] images into this one map. 1.5 m resolution real surface temperature published research [4:08] not a stock photo same place and same moment. Two completely different picture [4:17] of what actually happening on the ground. The gap is the whole top in one image. [4:29] So, hold the gap. So, here is the gap um in one image, right? So, AI detection at the top [4:38] and public understanding at the bottom. And between nothing. So, no after established [4:47] to stop for turning one into the other. This missing step is what motivates our research. [4:59] So, left side [5:03] What's it um what happens to environmental data today? [5:09] report PDF dead end. Nobody outside the lab over opens it. And if you look at left side and what [5:20] happens inside the framework? Yeah, satellite AI story experience and people. So, it doesn't end in [5:31] a file. It ends in a person who understand. Same starting point completely different destination. [5:46] So, what if um every AI generated environmental insight [5:52] ended it not in a PDF, but in public conversation? [6:01] So, how do you like close the gap? And then this research proposed a human-centered special [6:08] storytelling framework. Instead of stopping at AI generated insight, the framework translates [6:17] that insight into something people can understand and act on. It connects AI [6:27] um analysis with a narrative, special experience, and public engagement. [6:36] In other words, AI only tells us what's happening. [6:42] Yeah. But, knowing what is happening isn't enough. People still need to [6:48] understand the why it matters to them, and that's exactly like where the gap begins. [6:58] Translation. So, not better AI, better translation. [7:06] So, this framework has a five stages. It translates AI insight into human understanding. [7:15] The most important gap step is a human story. Let me show you uh what I mean. [7:28] So, this is where translation happens. On the left is what AI give us. Surface temperature, [7:38] tree canopy, heat map. It's accurate, but it doesn't tell people why they should care. [7:48] On the right, the data hasn't changed, but only meaning has. [7:54] Now we understand. So this is a where children like away for the like a school bus, [8:00] and then that's uh almost no shade, and like a planting tree could make this street a safer. [8:09] Something like this. Same data, and then different understandable. [8:19] So let's meet Sarah. So she lives in Phoenix. So she doesn't read the AI report. She just do [8:28] wants her neighborhood to be cooler. And then she doesn't need more AI, she needs understanding. [8:38] So her journey start from like [8:40] This journey start from like eight drones scan the Sarah's neighborhood. [8:47] AI identifies the hottest street, and then the Sarah still doesn't know what that means for her. [8:57] So this is a where storytelling comes in. So instead of seeing heat map, Sarah see sees her own [9:05] neighborhood. See She recognize familiar place, and understand how extreme heat affects daily [9:13] life. The data hasn't changed, only its meaning has. That's the moment understanding begins. [9:25] So meaning becomes action. Communities don't act on temperature maps, [9:32] and then they act on stories. They make the data meaningful. [9:41] So, this is is not a list of like benefits. It's a chain. Um transparence builds trust. [9:52] Trust encourage people to engage. And the engagement leads to better decisions. [9:59] And that's exactly what this framework is designed to do. Translate AI insight into [10:08] understanding. And understanding into action. [10:19] Come on. So, um this pretty AI generated intelligence and communication design [10:29] and to build the heart build like the heart barely spoken to each other. [10:35] And then the methodological curve and then it introduced the human-centered special storytelling [10:42] as a translation framework. Not just a decoration applied after fact. Practice. [10:53] It helps organizations transform AI output into public understanding. Turning what human-centered [11:03] AI means into something a team can actually act on. So, this is not another AI model. [11:11] This is translation framework that converts AI generated intelligence into human understanding. [11:31] Um before I I close, um, here I got four things you can do tomorrow, [11:40] and then go beyond the dashboard, design for understanding, [11:44] and bring designers into the AI process early, and the measure engagement, not just the model [11:55] act accuracy. So, because AI creates value only when people understand it well enough to act. [12:07] So, let me, um, end like a by looking ahead. Today, [12:13] special narrative transformation framework, SNTF, is a conceptual framework [12:22] built from interdisciplinary research. And then, my next step is to build a interactive prototype, [12:31] then turn AI-generated environmental intelligence into, uh, human-centered experience. And then, I [12:39] recently, uh, published a paper, my research, into a journal, and then I plan to do empirical, um, [12:49] like a analysis, and then, uh, develop like a further my like a previous research. [12:55] Then, I'll test like a whether it this approach see how people [13:00] better understand, engage with, and act on AI generate generated information. [13:07] Ultimately, I hope to apply and evaluate the framework in real-world [13:14] sustainability project. But, this future decision only the, um, one like a simple idea. [13:28] Today, AI can detect patterns faster and more accurately than ever before. [13:40] It already knows a great deal about our environment. [13:49] But knowledge alone doesn't change cities, [13:54] our environment. It doesn't change human behavior. It doesn't change community. [14:09] People do. When people understand and we care uh when they care, when we care, we can act. [14:21] So, that's the role of like a communication design and then special narrative uh [14:28] transformation framework. So, and that is why I believe that like the future of AI isn't [14:34] only about building better intelligence. It's about creating better understanding because [14:42] the intelligence has value only when people can understand it and connect with it and act on it. [14:52] That is the goal of the special narrative transformation framework. Thank you. [applause] [15:03] So, here is my like a contact information, email and LinkedIn. [15:08] Yeah. [laughter] So, if you're interested in like the SNT app [15:12] or like you want my research and then like you please contact me. Thank you very much.