Day 1: Maria Loureiro - AI Driven Damage Functions | ADIA Lab Symposium 2025

Day 1: Maria Loureiro - AI Driven Damage Functions | ADIA Lab Symposium 2025

🎙 Maria Loureiro 👥 824 📅 November 5, 2025 ⏱ 25 min 👁 34 📄 expert opinion 🧭 2026-08-16
Available in: English (current) Français

Keywords

damage functionsintegrated assessment modelssocial cost of carbonmachine learningclimate risk

Summary

Maria Loureiro, Professor of Economics at the University of Santiago de Compostela, presents her research on using AI to improve damage functions in climate-economic models. She critiques traditional linear and quadratic damage functions for underestimating climate impacts and ignoring extreme events and regional heterogeneity. She describes her work extending an integrated assessment model with 57 regions, testing various damage functions from the literature to estimate the social cost of carbon. Her findings show that the social cost of carbon is initially lower in high-income regions but increases sharply after 2040, becoming more homogeneous globally. She emphasizes the importance of incorporating extreme events and health impacts, citing data on heat-related deaths in Spain and Europe. She also discusses the role of misinformation in climate discourse, using social media data from the 2024 Dana floods in Spain to show how fake news spreads faster and hampers resilience. She concludes that AI can help better model climate impacts, both through numerical data and text analysis, to inform policy and risk assessment.

168 words

Critical Evaluation

Value of the Information & Strength of the Argument

The talk provides valuable insights into the limitations of current damage functions and the potential of AI to address them. The argumentation is coherent, moving from the problem of underestimation to specific examples and proposed solutions. The speaker supports her claims with references to her own research and general trends, but the presentation is more of an overview than a detailed exposition, leaving some arguments underdeveloped.

Scientific Rigor, Source Quality, Title Accuracy

The speaker references established models (e.g., Nordhaus, DICE) and mentions her own publications, but specific citations are not provided in the talk. The title accurately reflects the content. The talk is based on the speaker’s expertise and ongoing research, which adds credibility, but the lack of detailed references limits the ability to verify all claims.

136 words

Title / Content Match

The title accurately reflects the content, which focuses on AI-driven damage functions for climate policy and risk assessment.

Quality & Reliability

7/10

The speaker is a recognized expert in environmental economics, and the talk presents ongoing research with references to established models and empirical data. However, the presentation is a high-level overview without detailed methodology or peer-reviewed citations, limiting its immediate verifiability.

Key Moments

Cited Sources

Concurring Sources

Contribution & Novelties

The talk presents an ongoing research agenda that applies AI to improve damage functions in climate-economic models, addressing limitations of traditional approaches. It highlights the importance of incorporating extreme events, health impacts, and regional heterogeneity, and demonstrates the use of micro data and social media analysis. The speaker’s work on the social cost of carbon with a 57-region model provides new insights into temporal and regional variations.

Pour aller plus loin :

129 words

Radar Profile

The radar profile shows balanced scores across information quantity, quality, technical level, and reliability, with a slight emphasis on quality and reliability. This suggests a well-rounded presentation with solid content and credible sourcing, though not extremely technical or exhaustive.

Reliability 7/10