Comment on Hardening Firefox with Anthropic’s Red Team

• PabloSexcrowbar@piefed.social ⁨3⁩ ⁨weeks⁩ ago

  That’s…exactly my point though…

  source

  • lIlIlIlIlIlIl@lemmy.world ⁨3⁩ ⁨weeks⁩ ago

    What is?

    source

    • PabloSexcrowbar@piefed.social ⁨3⁩ ⁨weeks⁩ ago

      That even though the team is using AI to check for vulnerabilities, they’re trained and know when their AI is hallucinating and when it’s not.

      source

      • lIlIlIlIlIlIl@lemmy.world ⁨3⁩ ⁨weeks⁩ ago

        I guess I’m not sure how hallucinating and reading from source code are overlapping. Do you think these models are just barfing back garbage nonsense?

        source

        • -> View More Comments