Should You Trust Your Copilot? Limitations and Merits of AI Coding Assistants


AI-powered coding assistants, such as GitHub Copilot, are spreading rapidly in the software engineering community.

Copilot was developed by Microsoft and OpenAI on top of Codex, a transformer-based Large Language Model, and overtook 400.000 subscribers in the first month. It was praised by influential engineers, including Guido van Rossum, the inventor of the Python language [1, 2].

Several industrial and academic studies confirm that AI coding assistants yield an increment in productivity, but other works raise concerns about their widespread use [3].

The Free Software Foundation, an influential independent institution, complained about the training process of Codex, which allegedly violated the licensing of open-source code [4]. A study from the Association for Computing Machinery highlighted the danger of including coding assistants in education paths, concluding that an unrestricted adoption may lead to over-dependence and a decrease in learners' understanding [5]. Finally, in a recent preprint, researchers from Stanford describe a controlled experiment suggesting that coding assistants may lead to overconfidence and to an increase in security-related bugs [6].

By providing an unbiased and independent review of the literature, this talk aims to inform the debate about the trustworthiness of AI coding assistants and to provide insights into their future evolution.


[1] Satya Nadella. “[Fiscal Year 2022 – Fourth Quarter Earnings”. 2022]( Accessed on 8th February 2023.

[2] Guido van Rossum, Lex Friedman. “[Guido van Rossum on GitHub Copilot: It helps take care of the boring stuff](”. 2022, Accessed on 8th February 2023.

[3] Saki Imai. “Is GitHub copilot a substitute for human pair-programming? An empirical study.” Proceedings of the ACM/IEEE 44th International Conference on Software Engineering: Companion Proceedings. 2022.

[4] Craig Topham. “Publication of the FSF-funded white papers on questions around Copilot”. 2022. Accessed on 8th February 2023.

[5] James Finnie-Ansley et al. “The robots are coming: Exploring the implications of openai codex on introductory programming.” Australasian Computing Education Conference. 2022.

[6] Neil Perry et al. “Do Users Write More Insecure Code with AI Assistants?.” arXiv preprint arXiv:2211.03622 (2022).


Emanuele is Engineer by education, Data Scientist by choice, researcher and lecturer by passion. During his PhD in ML, he got invited to EPFL Lausanne for a 6-month visit and published 9 papers in top journals.He is the co-founder of xtream, an AI boutique applying academic research to business. Contributing to the community is part of their mission: He was a speaker and track organizer at eRum, AMLD, and PyCon and he lectured at Italian, Swiss, and Polish universities.

Open Data Science




Open Data Science
One Broadway
Cambridge, MA 02142

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from Youtube
Consent to display content from Vimeo
Google Maps
Consent to display content from Google