Research recitation - GitHub Docs

submited by

Style Pass

2021-06-30 14:30:10

GitHub Copilot helps you write code leveraging the collective intelligence of software developers worldwide. Copilot has already read through lots of public code, then it considers your own code, tries to guess what you want to do and comes up with a suggestion for how to get you there. That suggestion is based on your code. But indirectly, it’s informed by the code of those who came before you.

How direct is the relationship between the suggested code and the code that informed it? In a recent thought-provoking paper1, Bender, Gebru et al. coined the phrase “stochastic parrots” for artificial intelligence systems like the ones that power GitHub Copilot. Or as a fellow machine learning engineer at GitHub2 remarked during a water cooler chat: these systems can feel like ”a toddler with a photographic memory.”

These are deliberate oversimplifications. Many GitHub Copilot suggestions feel pretty specifically tailored to the particular code base the user is working on. Often, it looks less like a parrot and more like a crow building novel tools out of small blocks3. But there’s no denying that GitHub Copilot has an impressive memory: