I'm curious about how to identify the differences between code written by a person versus that generated by an AI. I'm considering starting a side hustle with an AI training company that focuses on reviewing code and want to know what signs to look for.
4 Answers
I've actually found AI code can be incomplete, like it's got a storyline that's missing parts—just like plot holes in a movie! It doesn't always follow existing functions or can be inconsistent in its patterns.
One clear sign is if the person submitting the code has no clue about how it works—chances are it was generated by AI. But honestly, I've seen plenty of coworkers push code they didn't understand long before AI came around.
Totally! This has definitely been a recurring issue, and it's not something new at all.
The comments in the code can give it away too. AI tends to produce comments with perfect grammar and often states the obvious—like things a human wouldn't find necessary to mention in detail.
You can often spot AI-generated code because it usually looks really polished but ends up running poorly. On the flip side, human code might look a bit messy but should execute just fine.
Interesting way to put it! What do you mean by 'plot holes' exactly?