IIPFCC Submits New Views to the UNFCCC Climate Finance Work Programme: A Climate-Resilience Proposal Card Model for Taiwan's 55 Indigenous Areas
Original Chinese title: IIPFCC向UNFCCC提交最新CFWP意見：55原鄉可把傳統知識、風險節點與行動證據接成氣候韌性提案資料卡
On September 8, the UNFCCC published a new International Indigenous Peoples’ Forum on Climate Change (IIPFCC) submission to the Climate Finance Work Programme (CFWP). For Taiwan's 55 Indigenous township areas, a practical next step is to organize local risk nodes, community-authorized traditional knowledge, action evidence, human verification and official service routes into maintainable proposal cards that RAG and chatbots can help index and update.
Yuan Media AI Editorial Desk
Yuan Media AI Editorial Desk follows public policy, local governance, Indigenous traditional knowledge, climate resilience, digital public services and AI data governance across Taiwan's Indigenous regions.
On September 8, 2026, the United Nations Framework Convention on Climate Change (UNFCCC) document system published an “International Indigenous Peoples’ Forum on Climate Change (IIPFCC) submission on the CFWP.” The public record identifies it as a submission by an external organization and places it under the topic of climate finance. The UNFCCC page for the Climate Finance Work Programme explains that the CFWP is a two-year 2026–2027 programme established by the CMA.7 decision and related to Article 9 of the Paris Agreement.
The most reliable information available directly from the public document page is that IIPFCC made a formal submission dated September 8 and that it concerns climate finance. This article therefore uses that fresh signal alongside established UNFCCC, Green Climate Fund (GCF), Local Communities and Indigenous Peoples Platform (LCIPP), CARE Principles and Local Contexts frameworks to ask a practical question: what information infrastructure can Taiwan's 55 Indigenous township areas prepare now for future climate-resilience work?
Start with traceable information and a maintainable small index
The GCF Indigenous Peoples Policy recognizes Indigenous Peoples as important partners in climate mitigation and adaptation and emphasizes meaningful engagement and respect for rights, lands, territories, natural resources and cultural resources. The UNFCCC LCIPP Resource Hub likewise brings together Indigenous and local-community knowledge, practices and climate-action resources. Together these frameworks point to a broader understanding of climate information: rainfall, temperature and satellite data matter, but so do long-term observations of water, crops, forests, fisheries, seasonal cycles, ceremonies and disaster experience. The rules governing whether such knowledge may be shared or reused should travel with the data itself.
A practical first step for Taiwan's Indigenous areas is a maintainable “climate-resilience proposal card.” Each card should address one clear local issue—for example drought backup water, farm-road disruption, crop heat stress, flash flooding, wildfire risk, coastal erosion, evacuation nodes, or intergenerational transmission of climate knowledge. Consistent fields make later connection to spreadsheets, GIS, RAG, chatbots and existing township-office systems much easier.
Four layers for each proposal card
The first layer is risk and place: location or service area, hazard type, affected people and livelihoods, current status, last human verification date, and links back to official weather, water, road or emergency information.
The second layer is traditional knowledge and permissions. If a card contains community water names, seasonal observations, ceremonial protocols, farming knowledge, gathering routes or oral histories, it should also record the source community, steward, access level, conditions of use, reuse status and withdrawal or re-confirmation process. The CARE Principles and Local Contexts Traditional Knowledge Labels provide useful reference points for designing these fields while leaving actual decisions with communities themselves.
The third layer is action and evidence: what has already been done, who is responsible, current status, photographs, sensor observations, maps, meeting records, human verification and outcome indicators. This can reduce the need to reconstruct fragmented evidence every time a community or township prepares a plan or coordinates across agencies.
The fourth layer is the official service route. It should list public programme names, intended users, official sources, application periods, contact points, current status and next steps. It should not determine eligibility or turn financial amounts into an accountability narrative. The purpose is to help residents and local staff find the correct official route more quickly.
Use RAG and chatbots for indexing, reminders and handoff
Within this model, RAG should read only approved public or authorized indexes. It can answer where information came from, when it was last updated, which public programmes are currently listed, which fields are missing and which human contact should handle the next step. A chatbot can translate complex notices into readable steps and flag expired sources or records that need human review.
The model should not decide whether a community qualifies for climate finance, whether traditional knowledge may be disclosed, or whether a local hazard meets a statutory disaster threshold. Those decisions remain with competent authorities, community rights holders and responsible emergency agencies. Limiting AI to finding, organizing, reminding and routing makes the system easier to govern over time.
A 90-day MVP is enough to test the workflow
In month one, one township could select two or three risk types and create 10–20 proposal cards. In month two, township staff, community organizations, agricultural workers and cultural knowledge holders can review the fields, permissions and human-verification rules. In month three, a lightweight RAG or chatbot can be connected to the approved index and tested for source retrieval, missing-field reminders and human handoff.
Useful early indicators include the share of records with traceable sources, the share still within their update period, the share of traditional-knowledge records with clear permissions, time needed to find the correct official contact, whether AI responses link back to source material, and whether users can reach a person when necessary. These foundations make future local climate proposals easier to explain without separating evidence from cultural rights.
Public sources
AI use and content-safety disclosure
This AI-assisted draft is based on public UNFCCC, Green Climate Fund, LCIPP, GIDA and Local Contexts materials. It does not attribute specific positions to IIPFCC beyond the metadata visible on the public document page and does not determine funding eligibility or legal rights.