April 24, 2024 | Raghu Raman, Hiran H. Lathabai, Santanu Mandal, Payel Das, Tavleen Kaur, Prema Nedungadi
This study evaluates the SDG literacy and intelligence of ChatGPT, a large language model (LLM), using two widely used sustainability assessment tools: the UN SDG Fitness Test and the Sustainability Literacy Test (SULITEST). The results indicate that ChatGPT demonstrates high SDG literacy, but its comprehensive SDG intelligence requires further exploration. The SDG Fitness Test assesses eight key competencies across introductory, intermediate, and advanced levels. However, both tests were found to be insufficient in mapping to the 17 SDGs and eight cross-cutting SDG core competencies. SULITEST could map only 5 out of 8 competencies, while the Fitness Test mapped 6 out of 8. Both tests fell short in covering the 17 SDGs, with many SDGs underrepresented or not represented at all. Consequently, both tools proved ineffective in assessing SDG intelligence through SDG coverage. The study recommends future versions of ChatGPT to enhance competencies such as collaboration, critical thinking, systems thinking, and others to achieve the SDGs. It concludes that while AI models like ChatGPT hold considerable potential in sustainable development, their usage must be approached carefully, considering current limitations and ethical implications. The study also highlights the need for further research on the potential of AI and ChatGPT in addressing SDGs, emphasizing the importance of ensuring that AI development and usage align with SDGs, promoting sustainability at every level.This study evaluates the SDG literacy and intelligence of ChatGPT, a large language model (LLM), using two widely used sustainability assessment tools: the UN SDG Fitness Test and the Sustainability Literacy Test (SULITEST). The results indicate that ChatGPT demonstrates high SDG literacy, but its comprehensive SDG intelligence requires further exploration. The SDG Fitness Test assesses eight key competencies across introductory, intermediate, and advanced levels. However, both tests were found to be insufficient in mapping to the 17 SDGs and eight cross-cutting SDG core competencies. SULITEST could map only 5 out of 8 competencies, while the Fitness Test mapped 6 out of 8. Both tests fell short in covering the 17 SDGs, with many SDGs underrepresented or not represented at all. Consequently, both tools proved ineffective in assessing SDG intelligence through SDG coverage. The study recommends future versions of ChatGPT to enhance competencies such as collaboration, critical thinking, systems thinking, and others to achieve the SDGs. It concludes that while AI models like ChatGPT hold considerable potential in sustainable development, their usage must be approached carefully, considering current limitations and ethical implications. The study also highlights the need for further research on the potential of AI and ChatGPT in addressing SDGs, emphasizing the importance of ensuring that AI development and usage align with SDGs, promoting sustainability at every level.