Last month GitHub released Copilot, an AI powered tool that provides surprising accurate suggestions ranging from a few lines of code to entire functions. Copilot uses a vast and powerful AI model that is trained on billions of lines of open source code from GitHub. This, combined with the context provided by your code and comments, allows it to provide amazingly accurate suggestions. This is no simple autocomplete, this is pair programming with a robot!

From my experiences so far, while Copilot has a real “wow” factor, there are some limitations, most notably the unpredictability of its suggestions and the cognitive load required to review the code it generates. This tool also opens up many interesting debates, from licencing concerns (is this really fair use of open source code? and what about attribution?), to the existential - are we going to be replaced by robots?

If you’re interested to see a glimpse of what AI-assisted programming might look like in the future, then come along. The robots are coming…