Well, this is just a whole bunch of "maybe". I think maybe we can train this further using existing linters and analyzers? At least the AI will emit far fewer

Well, this is just a whole bunch of "maybe". I think maybe we can train this further using existing linters and analyzers? At least the AI will emit far fewer lines of critically dangerous code, but we'll still have a lot of anti-pattern issues. Maybe GPT isn't really a good fit for this kind of task. Maybe we can create a better assistant using simpler AIs. If we reduce the scope (e.g. language, framework) and programming style (e.g. OOP, code organization, design), the amount of context should be much smaller than what GPT is hoarding. This may also allow us to have some open-source programming AIs.