4 ms·
Unless we get to a point where entire codebases can be maintained and extended with code that adheres to functional requirements and is free of bugs, there’s st
by heresjohnny 4y ago
Unless we get to a point where entire codebases can be maintained and extended with code that adheres to functional requirements and is free of bugs, there’s still some room for humans.
Consider highly contextual problems: today I faced an issue where some updated code could have negative side-effects given certain database state. A regular set of tests did not catch this issue – only my knowledge of the possible states did. I doubt an AI will be able to solve such problems any time soon.
- sothatsit 4y agoI agree that building systems that can keep track of large-scale programs seems very difficult, and the current LLMs don’t seem up to the task. However, I wonder whether we can leverage the ability of LLMs to understand short code snippets to inform higher-level static analysis. I think we may be able to generate high-level models of codebases using the intuition of LLMs, that we could then use to run higher-level simulations to detect possible bugs. It seems like it could be possible, but soon? I agree, probably not.