ChatGPT generates code, then warns about potential problems. author ignores problems, runs code as if there was not a single line of input validation in any of

ChatGPT generates code, then warns about potential problems. author ignores problems, runs code as if there was not a single line of input validation in any of their code, blames CGPT. that being said there are pretty bad edge cases, IIRC I asked it to explain some clojure function and I got ENTIRELY different answer when I changed input variable name from `event` to `events`, because CGPT assumed that context of naming implies how code works.