logoalt Hacker News

zombotyesterday at 5:23 AM0 repliesview on HN

Yes, when a human does it. With an LLM, the "trust me, I took extra care" is extremely doubtful. If LLMs could do that, they might as well use unsafe languages.