SOLVING PHYSICS PROBLEMS WITH THE HELP OF ARTIFICIAL INTELLIGENCE: AN ANALYSIS OF THE CHATGPT, GEMINI AND DEEPSEEK MODELS

Article Sidebar

PDF (Uzbek (Latin))
Published: 2025-06-05
DOI: https://doi.org/10.66073/researchfocus.v4i5.1604
Keywords:
Artificial intelligence, physics problems, ChatGPT, Gemini, DeepSeek, mechanics, molecular physics, educational technologies, large language models, comparative analysis

Main Article Content

Rasulov Shohruh Mo‘sinjon o‘g‘li
Farg‘ona davlat universiteti, fizika yo‘nalishi 4-kurs talabasi
Axmedov Baxodir Baxramovich
Associate professor of Fergana State University, f.-m.f.b.f.d.

Abstract

This article provides a comparative analysis of the effectiveness of the modern artificial intelligence models ChatGPT, Gemini and DeepSeek in solving problems in physics. In the study, 130 physics problems (100 in mechanics, 30 in molecular physics) corresponding to the higher education programmes of Uzbekistan were selected and given to each model to solve. The results showed that the success rate of the ChatGPT and DeepSeek models in the mechanics section amounted to 92%, whereas the result of the Gemini model was 85%. In the molecular physics section DeepSeek recorded a relatively higher indicator. The article analyses the errors made by the models and explains their causes. The results of the study identify the possibilities of using artificial intelligence in physics education and serve as an important basis for developing practical recommendations for teachers.

Article Details

Issue

Vol. 4 No. 5 (2025): Research Focus International Scientific Journal. Volume 4, Issue 5 (2025)

Section

01.00.00 – Physical and Mathematical Sciences
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

How to Cite

SOLVING PHYSICS PROBLEMS WITH THE HELP OF ARTIFICIAL INTELLIGENCE: AN ANALYSIS OF THE CHATGPT, GEMINI AND DEEPSEEK MODELS. (2025). Research Focus International Scientific Journal, 4(5), 8-15. https://doi.org/10.66073/researchfocus.v4i5.1604

References

Mirziyoyev, Sh. M. (2024). Sun’iy intellekt texnologiyalarini 2030-yilgacha rivojlantirish strategiyasini tasdiqlash to‘g‘risida. PQ-358-son qaror, 14-oktabr. O‘zbekiston Respublikasi Prezidenti qarori.

Shoaxmedova, N., & Yusupova, D. (2023). Ta’limda sun’iy intellekt texnologiyalarini qo‘llash imkoniyatlari. Raqamli iqtisodiyot va axborot texnologiyalari, 1(1), 36–42.

Mustafoyev, O. (2024). Sun’iy intellekt texnologiyalari asosida o‘quvchilarning o‘quv faoliyatini faollashtirishning shakl, metod va vositalari. Buxoro davlat pedagogika instituti jurnali, 4(4), 55–60. (Onlayn)

Liang, Y., Zou, D., Xie, H., & Wang, F. L. (2023). Exploring the potential of using ChatGPT in physics education. Smart Learning Environments, 10, Article 52. https://doi.org/10.1186/s40561-023-00273-7

Wang, K. D., Burkholder, E., Wieman, C., Salehi, S., & Haber, N. (2023). Examining the potential and pitfalls of ChatGPT in science and engineering problem-solving. arXiv preprint, arXiv:2310.08773 [cs.AI].

OpenAI. (2023). GPT-4 Technical Report. arXiv preprint, arXiv:2303.08774 [cs.CL].

Gemini Team. (2024). Gemini: A family of highly capable multimodal models. arXiv preprint, arXiv:2312.11805 [cs.CL].

Brodsky, S. (2025, February 4). DeepSeek shows there’s room for more AI players. IBM News. (Onlayn)

Mamatova, G., & Ismoilov, H. (2024). Mustaqil ta’limda sun’iy intellektdan foydalanish. Universal xalqaro ilmiy jurnal, 1(12), 599–602. (Onlayn)