We ahve numerous, well-doocumented cases of LLMs being just plain wrong and/or hallucinating. There is an epidemic of LLM usage in college (and even high school

We ahve numerous, well-doocumented cases of LLMs being just plain wrong and/or hallucinating. There is an epidemic of LLM usage in college (and even high school) to the point where lack of critical thinking is becoming a massive problem and will only get worse. This represents a huge security threat too if code is uncritically applied to code bases. We've seen many examples where people try and influence LLM output (eg [1][2]). These attempts have generally varied from crude to laughably bad but they'll get better. Is it so hard to imagine prompt injection being a serious security threat to a code base? That aside, I just don't understand being "all in" on LLMs for coding. Is this really productive? How much boilerplate do you actually write? With good knowledge of a language or framework, that tends to be stuff you can spit out really quickly anyway. [1]: https://www.theguardian.com/technology/2025/may/16/elon-musks-ai-firm-blames-unauthorised-change-for-chatbots-rant-about-white-genocide https://www.theguardian.com/technology/2025/may/16/elon-musk... [2]: https://www.theguardian.com/technology/2025/jan/28/we-tried-out-deepseek-it-works-well-until-we-asked-it-about-tiananmen-square-and-taiwan https://www.theguardian.com/technology/2025/jan/28/we-tried-...