{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T05:55:27Z","timestamp":1792130127743,"version":"4.4.0"},"reference-count":32,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T00:00:00Z","timestamp":1792108800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Artif. Intell."],"abstract":"<jats:p>We investigate whether trade openness, governance, environmental taxation and selected macroeconomic indicators can classify the contemporaneous CO2 emission level of countries observed over 2000 to 2023, using data from the World Development Indicators and the OECD. CO2 emissions per capita are divided into three ordinal levels, low, medium and high, using tertile cut points estimated on the training data only within every validation split. Four classifiers, Random Forest, XGBoost, a support vector machine and a multinomial logistic regression, are tuned and compared alongside two models designed specifically for ordinal outcomes, an ordered logistic regression and an ordinal decomposition of XGBoost, under four validation designs: a country grouped holdout and country grouped 5 fold cross validation, in which no country appears in both the training and test sets; a chronological holdout; a 1 year ahead forecasting task; and a random stratified country year split, retained only as a reference point. Under country grouped cross validation, no classifier achieves a decisively higher accuracy than the others: nested cross validation gives a mean accuracy of 0.644 for XGBoost, 0.648 for Random Forest, 0.671 for the support vector machine, 0.674 for the multinomial logit, 0.676 for the ordered logit and 0.637 for the ordinal XGBoost model, with overlapping confidence intervals throughout. This is markedly lower than the 0.91 accuracy obtained under a random country year split with the same tuned pipeline, a gap of 12.5 percentage points for XGBoost, confirming that a substantial part of the high accuracy obtainable under random splitting reflects the model recognizing previously seen country profiles rather than generalizing to unseen countries. Accuracy falls less sharply under the chronological holdout and the 1 year ahead forecasting task, in the region of 0.65 to 0.74, indicating that the classifiers generalize better across time within a country than across countries. Permutation importance, SHAP and LIME consistently identify government effectiveness, urban population growth and environmental tax revenue as the features most strongly associated with model predictions, with a moderate contribution from trade openness and the smallest from GDP growth, a ranking that is stable across importance methods, learning algorithms and cross validation folds. These variables are predictors of the model's output, not estimated causal determinants of emissions, and the results are best read as evidence for climate risk monitoring and further policy evaluation rather than as a causal policy assessment.<\/jats:p>","DOI":"10.3389\/frai.2026.1943572","type":"journal-article","created":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T05:40:14Z","timestamp":1792129214000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["A global machine learning framework for classifying national CO2 emission levels: responsible and explainable artificial intelligence for sustainable climate governance"],"prefix":"10.3389","volume":"9","author":[{"given":"Muhammad Ali","family":"Khawaja","sequence":"first","affiliation":[{"name":"Applied Economics Research Centre, University of Karachi","place":["Karachi, Pakistan"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Muhammad Farooq","family":"Shaikh","sequence":"additional","affiliation":[{"name":"School of Engineering, University of Warwick","place":["Coventry, United Kingdom"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,10,16]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"48424","DOI":"10.1007\/s11356-022-18711-3","article-title":"Influencing factors of carbon emissions and their trends in China and India: a machine learning method","volume":"29","author":"Ahmed","year":"2022","journal-title":"Environ. Sci. Pollut. Res."},{"key":"B2","doi-asserted-by":"publisher","first-page":"2581","DOI":"10.1007\/s00146-025-02615-0","article-title":"AI and climate: an ethical sustainability framework for balancing risks and responsibilities","volume":"41","author":"Amore","year":"2026","journal-title":"AI Soc."},{"key":"B3","doi-asserted-by":"publisher","first-page":"15919","DOI":"10.1007\/s10668-024-05566-6","article-title":"Determinants of carbon emissions in Africa: new evidence based on machine learning algorithms","volume":"28","author":"Ayimadu","year":"2024","journal-title":"Environ. Dev. Sustain."},{"key":"B4","doi-asserted-by":"publisher","first-page":"19469","DOI":"10.1038\/s41598-025-04236-5","article-title":"A machine learning approach to carbon emissions prediction of the top eleven emitters by 2030 and their prospects for meeting Paris agreement targets","volume":"15","author":"Begum","year":"2025","journal-title":"Sci. Rep."},{"key":"B5","doi-asserted-by":"publisher","first-page":"7581","DOI":"10.1007\/s11135-026-02594-4","article-title":"Multidimensional analysis of global economic freedom performances: structural findings and machine learning approaches","volume":"60","author":"Celik","year":"2026","journal-title":"Qual. Quant."},{"key":"B6","doi-asserted-by":"publisher","first-page":"1491","DOI":"10.3390\/math14091491","article-title":"Beyond the black box: nonlinear regimes and explainable AI in global innovation systems","volume":"14","author":"Celik","year":"","journal-title":"Mathematics"},{"key":"B7","doi-asserted-by":"publisher","DOI":"10.1177\/08944393261419809","article-title":"Global gender inequality through explainable AI: machine learning, clustering and SHAP insights","author":"Celik","year":"2026","journal-title":"Soc. Sci. Comput. Rev"},{"key":"B8","doi-asserted-by":"publisher","first-page":"7411","DOI":"10.3390\/su17167411","article-title":"Global sustainability performance and regional disparities: a machine learning approach based on the 2025 SDG Index","volume":"17","author":"Celik","year":"2025","journal-title":"Sustainability"},{"key":"B9","doi-asserted-by":"publisher","DOI":"10.1007\/s11135-026-02801-2","article-title":"Mapping latent environmental performance regimes: a spatial and explainable machine learning approach","author":"Celik","year":"2026","journal-title":"Qual. Quant"},{"key":"B10","doi-asserted-by":"publisher","first-page":"24","DOI":"10.1007\/s10101-026-00365-3","article-title":"Unveiling structural patterns of global budget transparency using advanced machine learning and explainable artificial intelligence frameworks","volume":"27","author":"Celik","year":"","journal-title":"Econ. Govern."},{"key":"B11","doi-asserted-by":"publisher","first-page":"273","DOI":"10.1023\/A:1022627411411","article-title":"Support vector networks","volume":"20","author":"Cortes","year":"1995","journal-title":"Mach. Learn."},{"key":"B12","article-title":"\u201cMild cognitive impairment: data driven prediction, risk factors and workup,\u201d","volume-title":"AMIA Summits on Translational Science Proceedings","author":"Fouladvand","year":"2023"},{"key":"B13","doi-asserted-by":"publisher","first-page":"145","DOI":"10.1007\/3-540-44795-4_13","article-title":"\u201cA simple approach to ordinal classification,\u201d","author":"Frank","year":"2001","journal-title":"Proceedings of the 12th European conference on machine learning"},{"key":"B14","doi-asserted-by":"publisher","first-page":"965","DOI":"10.5194\/essd-17-965-2025","article-title":"Global carbon budget 2024","volume":"17","author":"Friedlingstein","year":"2025","journal-title":"Earth Syst. Sci. Data"},{"key":"B15","doi-asserted-by":"crossref","first-page":"213","DOI":"10.4018\/979-8-2600-2020-3.ch007","article-title":"\u201cClimate intelligence integration: machine learning, governance and sustainable development pathways,\u201d","volume-title":"Responsible AI for Climate Action: Governance, Justice and Sustainable Transitions","author":"Gavali","year":"2026"},{"key":"B16","year":"2021","journal-title":"Net Zero by 2050"},{"key":"B17","volume-title":"Implementation of Various Machine Learning Algorithms for Traffic Sign Detection and Recognition","author":"Kalange","year":"2022"},{"key":"B18","volume-title":"IPCC, 2023: Climate Change 2023: Synthesis Report: A Report of the Intergovernmental Panel on Climate Change","author":"Lee","year":"2023"},{"key":"B19","doi-asserted-by":"publisher","first-page":"721517","DOI":"10.3389\/fenvs.2021.721517","article-title":"Driving factors of CO2 emissions: further study based on machine learning","volume":"9","author":"Li","year":"2021","journal-title":"Front. Environ. Sci."},{"key":"B20","doi-asserted-by":"publisher","first-page":"104276","DOI":"10.1016\/j.trd.2024.104276","article-title":"Estimation of transport CO2 emissions using machine learning algorithm","volume":"133","author":"Li","year":"2024","journal-title":"Transp. Res. Part D: Transport Environ."},{"key":"B21","doi-asserted-by":"publisher","first-page":"56","DOI":"10.1038\/s42256-019-0138-9","article-title":"From local explanations to global understanding with explainable AI for trees","volume":"2","author":"Lundberg","year":"2020","journal-title":"Nat. Mach. Intell."},{"key":"B22","article-title":"A unified approach to interpreting model predictions","author":"Lundberg","year":"2017","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B23","doi-asserted-by":"crossref","first-page":"135","DOI":"10.4018\/979-8-2600-2020-3.ch005","article-title":"\u201cArtificial intelligence for climate action and sustainable transitions: conceptual frameworks, governance models and responsible innovation pathways,\u201d","author":"Mishra","year":"2026","journal-title":"Responsible AI for Climate Action: Governance, Justice and Sustainable Transitions"},{"key":"B24","doi-asserted-by":"publisher","first-page":"218","DOI":"10.1016\/j.spc.2022.06.027","article-title":"The estimation of the carbon dioxide emission and driving factors in China based on machine learning methods","volume":"33","author":"Qin","year":"2022","journal-title":"Sustain. Prod. Consumpt."},{"key":"B25","doi-asserted-by":"publisher","first-page":"81","DOI":"10.1007\/s43681-025-00909-z","article-title":"Responsible intelligence: ethical AI governance for climate prediction in the Australian context","volume":"6","author":"Randeniya","year":"2026","journal-title":"AI Ethics"},{"key":"B26","doi-asserted-by":"publisher","first-page":"1135","DOI":"10.18653\/v1\/N16-3020","article-title":"\u201cWhy should I trust you? Explaining the predictions of any classifier,\u201d","author":"Ribeiro","year":"2016","journal-title":"Proceedings of the 22nd ACM SIGKDD international conference on knowledge discovery and data mining"},{"key":"B27","doi-asserted-by":"publisher","first-page":"511","DOI":"10.32479\/ijeep.19622","article-title":"Antecedents and forecasting of carbon emissions using machine learning algorithms: insights from the top ten carbon emitting nations","volume":"15","author":"Selmeya","year":"2025","journal-title":"Int. J. Energy Econ. Policy"},{"key":"B28","year":"1997","journal-title":"Kyoto Protocol to the United Nations Framework Convention on Climate Change"},{"key":"B29","year":"2015","journal-title":"Paris Agreement"},{"key":"B30","year":"2024","journal-title":"Emissions Gap Report 2024: No More Hot Air, Please, With a Massive Gap Between Rhetoric and Reality, Countries Draft New Climate Commitments"},{"key":"B31","year":"2023","journal-title":"State of the Global Climate 2023"},{"key":"B32","doi-asserted-by":"publisher","first-page":"23609","DOI":"10.1038\/s41598-024-75753-y","article-title":"A machine learning algorithm to explore the drivers of carbon emissions in Chinese cities","volume":"14","author":"Yu","year":"2024","journal-title":"Sci. Rep."}],"container-title":["Frontiers in Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1943572\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,10,16]],"date-time":"2026-10-16T05:40:16Z","timestamp":1792129216000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frai.2026.1943572\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10,16]]},"references-count":32,"alternative-id":["10.3389\/frai.2026.1943572"],"URL":"https:\/\/doi.org\/10.3389\/frai.2026.1943572","relation":{},"ISSN":["2624-8212"],"issn-type":[{"value":"2624-8212","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10,16]]},"article-number":"1943572"}}