Thread

1. intrepidsoldier · 1790914300 · focus · HN ↗

   Just the beginning. AI is going to expose how fragile the entire computing infrastructure in our world is.

   1. ankurdhama · 1790914599 · focus · HN ↗

      Does this also mean the code generated and reviewed by LLMs will not have such issues going forward?

      1. trollbridge · 1790914892 · focus · HN ↗

         Quite the opposite, particularly when the biggest vendors of coding agents insist on not allowing their models to be used to check the code they generate for security issues.

         1. enraged_camel · 1790916896 · focus · HN ↗

            I've been able to use Opus 5.5 and Fable 5.1 for defensive security audits without any issues. They cannot do offensive tasks like pen-testing but in a lot of cases that's not a big shortcoming.

            1. whiskey-one · 1790945426 · focus · HN ↗

               Any tips / online resources how to best utilize for defensive reviews?
            2. trollbridge · 1790954036 · focus · HN ↗

               I've repeatedly slammed into walls doing very basic tasks. It can do a dumbed-down security audit, but fails to do offensive tasks against my own codebase which is frankly how you use a model like this effectively.

Unofficial Hacker News client; not affiliated with Y Combinator.