Leveraging LLMs for predictive insights in food policy and behavioral interventions

Abstract Food consumption and production significantly contribute to global greenhouse gas emissions, making them key targets for climate change mitigation. Over the past two decades, food policy initiatives have focused on reshaping production and consumption patterns by reducing food waste and cur...

Full description

Saved in:
Bibliographic Details
Main Authors: Micha Kaiser, Paul M. Lohmann, Peter Ochieng, Billy Shi, Cass R. Sunstein, Lucia A. Reisch
Format: Article
Language:English
Published: Springer 2025-08-01
Series:Discover Food
Subjects:
Online Access:https://doi.org/10.1007/s44187-025-00552-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849333215509086208
author Micha Kaiser
Paul M. Lohmann
Peter Ochieng
Billy Shi
Cass R. Sunstein
Lucia A. Reisch
author_facet Micha Kaiser
Paul M. Lohmann
Peter Ochieng
Billy Shi
Cass R. Sunstein
Lucia A. Reisch
author_sort Micha Kaiser
collection DOAJ
description Abstract Food consumption and production significantly contribute to global greenhouse gas emissions, making them key targets for climate change mitigation. Over the past two decades, food policy initiatives have focused on reshaping production and consumption patterns by reducing food waste and curbing ruminant meat consumption. While evidence on effective interventions is improving, assessing appropriate and context-specific policies remains difficult due to external validity concerns. This paper demonstrates that a fine-tuned large language model (LLM) can accurately predict outcome directions in approximately 80% of empirical studies evaluating dietary interventions. Predictive accuracy improves with richer input detail, peaking at around 75 prompts before declining due to overfitting or saturation. To contextualize these results, we benchmark the LLM against both classical random-effects meta-regression and a prompt-based variant executed entirely within the model. Although traditional approaches yield reasonable magnitude estimates, they lag behind LLMs in directional accuracy and adaptability to diverse intervention formats. Together, our findings suggest that LLMs-especially when fine-tuned on curated evidence-offer a scalable pathway for data-driven, context-sensitive food policy modeling.
format Article
id doaj-art-74eac22d44254ce18b6fa040cbab3b4c
institution Kabale University
issn 2731-4286
language English
publishDate 2025-08-01
publisher Springer
record_format Article
series Discover Food
spelling doaj-art-74eac22d44254ce18b6fa040cbab3b4c2025-08-20T03:45:56ZengSpringerDiscover Food2731-42862025-08-015112510.1007/s44187-025-00552-xLeveraging LLMs for predictive insights in food policy and behavioral interventionsMicha Kaiser0Paul M. Lohmann1Peter Ochieng2Billy Shi3Cass R. Sunstein4Lucia A. Reisch5El-Erian Institute of Behavioural Economics and Policy, Cambridge Judge Business School, University of CambridgeEl-Erian Institute of Behavioural Economics and Policy, Cambridge Judge Business School, University of CambridgeYNOT Institute, Queens’ College, Cambridge, University of CambridgeYNOT Institute, Queens’ College, Cambridge, University of CambridgeHarvard Law School, Harvard UniversityEl-Erian Institute of Behavioural Economics and Policy, Cambridge Judge Business School, University of CambridgeAbstract Food consumption and production significantly contribute to global greenhouse gas emissions, making them key targets for climate change mitigation. Over the past two decades, food policy initiatives have focused on reshaping production and consumption patterns by reducing food waste and curbing ruminant meat consumption. While evidence on effective interventions is improving, assessing appropriate and context-specific policies remains difficult due to external validity concerns. This paper demonstrates that a fine-tuned large language model (LLM) can accurately predict outcome directions in approximately 80% of empirical studies evaluating dietary interventions. Predictive accuracy improves with richer input detail, peaking at around 75 prompts before declining due to overfitting or saturation. To contextualize these results, we benchmark the LLM against both classical random-effects meta-regression and a prompt-based variant executed entirely within the model. Although traditional approaches yield reasonable magnitude estimates, they lag behind LLMs in directional accuracy and adaptability to diverse intervention formats. Together, our findings suggest that LLMs-especially when fine-tuned on curated evidence-offer a scalable pathway for data-driven, context-sensitive food policy modeling.https://doi.org/10.1007/s44187-025-00552-xBehavioral public policyFood policyFood wasteLarge language models
spellingShingle Micha Kaiser
Paul M. Lohmann
Peter Ochieng
Billy Shi
Cass R. Sunstein
Lucia A. Reisch
Leveraging LLMs for predictive insights in food policy and behavioral interventions
Discover Food
Behavioral public policy
Food policy
Food waste
Large language models
title Leveraging LLMs for predictive insights in food policy and behavioral interventions
title_full Leveraging LLMs for predictive insights in food policy and behavioral interventions
title_fullStr Leveraging LLMs for predictive insights in food policy and behavioral interventions
title_full_unstemmed Leveraging LLMs for predictive insights in food policy and behavioral interventions
title_short Leveraging LLMs for predictive insights in food policy and behavioral interventions
title_sort leveraging llms for predictive insights in food policy and behavioral interventions
topic Behavioral public policy
Food policy
Food waste
Large language models
url https://doi.org/10.1007/s44187-025-00552-x
work_keys_str_mv AT michakaiser leveragingllmsforpredictiveinsightsinfoodpolicyandbehavioralinterventions
AT paulmlohmann leveragingllmsforpredictiveinsightsinfoodpolicyandbehavioralinterventions
AT peterochieng leveragingllmsforpredictiveinsightsinfoodpolicyandbehavioralinterventions
AT billyshi leveragingllmsforpredictiveinsightsinfoodpolicyandbehavioralinterventions
AT cassrsunstein leveragingllmsforpredictiveinsightsinfoodpolicyandbehavioralinterventions
AT luciaareisch leveragingllmsforpredictiveinsightsinfoodpolicyandbehavioralinterventions