NEW SCIENTIST: DeepMind has made software-writing AI that rivals average human coder

Press/Media: Expert comment

Period2 Feb 2022

Media contributions

1

Media contributions

  • TitleDeepMind has made software-writing AI that rivals average human coder
    Media name/outletNew Scientist
    Media typeWeb
    Country/TerritoryUnited Kingdom
    Date2/02/22
    DescriptionAlphaCode, like Copilot, was first trained on publicly available code hosted on GitHub. It was then fine-tuned on code from programming competitions. DeepMind says that AlphaCode doesn’t copy code from previous examples. Given the examples DeepMind provided in its preprint paper, it does appear to solve problems while only copying slightly more code from training data than humans already do, says Riza Theresa Batista-Navarro at the University of Manchester, UK.

    But AlphaCode seems to have been so finely tuned to solve complex challenges that the previous state of the art in AI coding tools can still outperform it on simpler tasks, she says.

    “What I noticed is that, while AlphaCode is able to do better than state-of-the-art AIs like GPT on the competition challenges, it does comparatively poorly on the introductory challenges,” says Batista-Navarro. “The assumption is that they wanted to do competition-level programming problems, to tackle more challenging programming problems rather than introductory ones. But this seems to show that the model was fine-tuned so well on the more complicated problems that, in a way, it’s kind of forgotten the introductory level problems.”
    URLhttps://www.newscientist.com/article/2306820-deepmind-has-made-software-writing-ai-that-rivals-average-human-coder/
    PersonsRiza Theresa Batista-Navarro

Keywords

  • Artificial intelligence
  • coding
  • computer science