Breach of high-profile accounts raises concerns about reliance on AI for security measures such as passwords
  • naught@sh.itjust.worksEnglish
    2·
    2 days ago

    You can do this and still use it to generate insecure code or even malicious code! Humans suck at reviewing

    • boonhet@sopuli.xyzEnglish
      21·
      1 day ago

      Unironically not a bad idea to have LLMs review eachothers code before committing. And then human review the PR.