POTENTIALS AND CHALLENGES OF GITHUB COPILOT AS AN ARTIFICIAL INTELLIGENCE TOOL
DOI:
https://doi.org/10.21728/p2p.2024v10n2e-7031Keywords:
Programmers, Developers, Tool, Codes, ProductivityAbstract
This article investigates the main potentialities and challenges of using GitHub Copilot as an Artificial Intelligence tool, highlighting the areas in which its benefits have been applied. This proposal was achieved based on the exploration and analysis of 12 empirical studies using GitHub Copilot as a tool used by Programmers/Developers, whose source was the repository of the Computer Science Center at Cornell University. The results showed that GitHub Copilot is a promising tool for increasing productivity, with a view to making the daily tasks of its users easier and more agile, so that its platform hosts a significant mass of users, made up of Programmers, Developers, Organizations and Code repositories. The study concludes that GitHub Copilot has potential to: increase Productivity, Code Programming, Natural Language Processing; make coding more efficient through new resources, the solution of algorithm and buggy problems, and these benefits occur in the area of Software Engineering, without exception. The challenges of using this tool are due to the Security of the generated code and the need to improve processes related to the tasks of Programmers/Developers. These diagnosed obstacles can be minimized through strategic tools available to mainly correct coding errors, and others, by the users themselves, in favor of the growth of GitHub Copilot.
Downloads
References
ANDRADE, R. O. ChatGPT reacende debate sobre o potencial criativo de sistemas de linguagem natural e as implicações éticas relacionadas ao seu uso. Universo Expandido da Inteligência Artificial. Revista Pesquisa FAPES, São Paulo, Edição 325, p. 17-22, mar. 2023. Disponível em: https://revistapesquisa.fapesp.br/wp-content/uploads/2023/03/016-022_capa-chatgpt_325-Parte-1.pdf. Acesso em: 20 abr. 2024.
BOSTROM, N. Superinteligência: Caminhos, perigos, estratégias. Rio de Janeiro: DarkSide Books, 2018. Disponível em: https://livrariapublica.com.br/livros/superinteligencia-caminhos-perigos-estrategias-nick-bostrom/. Acesso em: 20 abr. 2024.
CAETANO, L. M. D. Inteligência Artificial e ChatGPT: ameaça ou ferramenta na educação superior? 2023. Disponível em: https://www.researchgate.net/publication/370491136_Inteligencia_Artificial_e_ChatGPT_ameaca_ou_ferramenta_na_educacao_superior. Acesso em: 10 mar. 2024.
CHATTERJEE, S. et al. The impact of ai tool on engineering at anz bank an emperical study on github copilot within coporate environment. 2023. Disponível em: https://arxiv.org/ftp/arxiv/papers/2402/2402.05636.pdf. Acesso em: 10 mar. 2024.
DAKHEL, A. M. et al. GitHub Copilot AI pair programmer: Asset or Liability? arXiv:2206.15331v2 [cs.SE], p. 1-27, 14 Apr 2023. Disponível em: https://arxiv.org/pdf/2206.15331.pdf. Acesso em: 10 mar. 2024.
DOMINGOS, P. O algoritmo mestre. São Paulo: Novatec, 2017. Disponível em: https://pdfshared.com/download/B01N39DSJO?type=pdf&font=livros. Acesso em: 20 abr. 2024.
ERHABOR, D. et al. Measuring the Runtime Performance of Code Produced with GitHub Copilot. ArXiv:2305.06439v1 [cs.SE], p. 1-11, 10 May 2023. Disponível em: https://arxiv.org/pdf/2305.06439.pdf. Acesso em: 10 mar. 2024.
FU, Y. et al. Security Weaknesses of Copilot Generated Code in GitHub. ACM Trans. Softw. Eng. Methodol, v. 1, n. 1, p. 1-23, Apr. 2024. Disponível em: https://arxiv.org/pdf/2310.02059. Acesso em: 27 abr. 2024.
JAWORSKI, M.; PIOTRKOWSKI, D. Study of software developers’ experience using the Github Copilot Tool in the software development process. ArXiv:2301.04991v1 [cs.SE], p. 1-5, 12 jan. 2023. Disponível em: https://arxiv.org/pdf/2301.04991.pdf. Acesso em: 10 mar. 2024.
KOYANAGI, K. et al. Exploring the Effect of Multiple Natural Languages on Code Suggestion Using GitHub Copilot. Lisbon, v. 1, 2 fev. 2024. Disponível em: https://arxiv.org/pdf/2402.01438.pdf. Acesso em: 10 mar. 2024.
MATOS, C. GitHub Copilot: Como a ferramenta baseada em IA pode melhorar a produtividade do desenvolvedor. 2023. Disponível em: https://medium.com/the-code-career/github-copilot-como-a-ferramenta-baseada-em-ia-pode-melhorar-a-produtividade-do-desenvolvedor-a2a762ef37c6. Acesso em: 10 mar. 2024.
MAJDINASAB, V. et al. Assessing the Security of GitHub Copilot’s Generated Code: Targeted Replication Study. arXiv:2311.11177v1 [cs.SE], p. 1-10, 18 nov. 2023. Disponível em: https://arxiv.org/pdf/2311.11177v1. Acesso em: 10 mar. 2024.
MINAYO, M. C. S. (org.). Pesquisa social: teoria, método e criatividade. 18 ed. Petrópolis: Vozes, 2001. Disponível em: https://www.faed.udesc.br/arquivos/id_submenu/1428/minayo__2001.pdf. Acesso em: 20 abr. 2024.
MCCARTHY, J. What is artificial intelligence? Stanford University, nov. 2007. Disponível em: http://www-formal.stanford.edu/jmc/whatisai.pdf. Acesso em: 10 mar. 2024.
OLIVEIRA, A. Inteligência Artificial. Lisboa: Fundação Francisco Manuel dos Santos, 2019. Disponível em: https://ffms.pt/pt-pt/livraria/inteligencia-artificial. Acesso em: 10 abr. 2024.
POLSON, N.; SCOTT, J. Inteligência Artificial. Amadora: Vogais, 2020. Disponível em: https://www.wook.pt/livro/inteligencia-artificial-nick-polson/23894844. Acesso em: 10 abr. 2024.
PRODANOV, C. C.; FREITAS, E. C. Metodologia do trabalho científico: métodos e técnicas da pesquisa e do trabalho acadêmico. Novo Hamburgo, RS: Feevale, 2013. Disponível em: https://www.feevale.br/Comum/midias/0163c988-1f5d-496f-b118-a6e009a7a2f9/E-book%20Metodologia%20do%20Trabalho%20Cientifico.pdf. Acesso em: 10 mar. 2024.
RES, J. et al. Enhancing Security of AI-Based Code Synthesis with GitHub Copilot via Cheap and Efficient Prompt-Engineering. ArXiv:2403.12671v1 [cs.CR], 19 mar. 2024. Disponível em: https://arxiv.org/pdf/2403.12671.pdf. Acesso em: 10 abr. 2024.
RUSSEL, S.; NORVIG, P. Inteligência Artificial. 3. ed. Rio de Janeiro: Elsevier, 2013. Disponível em: https://1library.org/document/q7lj7nry-inteligencia-artificial-ed-russell-stuart-norvig-peter-pdf.html. Acesso em: 10 mar. 2024.
SICHMAN, J. S. Inteligência Artificial e sociedade: avanços e riscos. Estudos Avançados, São Paulo, v. 35, n. 101, p. 37-49, dez. 2021. Disponível em: https://www.revistas.usp.br/eav/article/view/185024/171207. Acesso em: 12 abr. 2024.
SOUZA, I. Inteligência artificial nas empresas: como aplicar e qual é o impacto. 2023. ZapSign. 28 fev. 2023. Disponível em: https://blog.zapsign.com.br/inteligencia-artificial-empresas/. Acesso em: 10 mar. 2024.
WERMELINGER, M. Using GitHub Copilot to Solve Simple Programming Problems. In: SIMPÓSIO TÉCNICO ACM DE EDUCAÇÃO EM CIÊNCIA DA COMPUTAÇÃO, 54., 2023. Proceeding [...]. Toronto: SIGCSE, 2023. p. 172-178. Doi: https://doi.org/10.1145/3545945.3569830. Disponível em: https://www.researchgate.net/publication/369248394_Using_GitHub_Copilot_to_Solve_Simple_Programming_Problems. Acesso em: 12 abr. 2024.
XU, H. GitHub Copilot: uma ferramenta inovadora para preencher automático de códigos. Universiti Utara Malaysia. dezembro de 2023. Disponível em: https://www.researchgate.net/publication/376406939_Github_Copilot_-_A_Groundbreaking_Code_Autocomplete_Tool. Acesso em: 10 mar. 2024.
YETIŞTIREN, B. et al. Evaluating the Code Quality of AI-Assisted Code Generation Tools: An Empirical Study on GitHub Copilot, Amazon CodeWhisperer, and ChatGPT. arXiv:2304.10778v2 [cs.SE], p. 1-45, 22 Oct 2023. Disponível em: https://arxiv.org/pdf/2304.10778.pdf. Acesso em: 10 mar. 2024.
ZHANG, B. et al. Practices and Challenges of Using GitHub Copilot: An Empirical Study. ArXiv:2303.08733v3 [cs.SE], p. 1-6, 27 Apr 2023a. Disponível em: https://arxiv.org/pdf/2303.08733.pdf. Acesso em: 28 abr. 2024.
ZHANG, B. et al. Demystifying Practices, Challenges and Expected Features of Using GitHub Copilot. arXiv:2309.05687v1 [cs.SE], p. 1-19, 13 Sep 2023b. Disponível em: https://arxiv.org/pdf/2309.05687.pdf. Acesso em: 10 mar. 2024.
ZHANG, D. et al. Artificial Intelligence Index Report 2022. Stanford University, March 2022. 230 p. Disponível em: https://arxiv.org/pdf/2205.03468. Acesso em: 10 mar. 2024.
ZHOU, X. et al. On the Concerns of Developers When Using GitHub Copilot. ArXiv:2311.01020v2 [cs.SE], p. 1-19, 28 Apr. 2024. Disponível em: https://arxiv.org/pdf/2311.01020. Acesso em: 29 abr. 2024.
Downloads
Published
Issue
Section
License
Copyright (c) 2024 Jefferson Sena, Adriano Barreto, John Barbosa, Keyla Alves
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
The journal is published under the Creative Commons - Attribution - Noncommercial - Share Alike 3.0 Brazil.
The published work is considered collaboration and therefore the author will not receive any remuneration for this as well as anything will be charged in exchange for publication.
All texts are responsibility of the authors.
It’s allowed partial or total reproduction of the texts of the magazine since the source is cited.
