Cyber evals

Week 44 · 26 Oct-1 Nov

1 source filed · EN 1

Week 43 · 19-25 Oct

2 sources filed · EN 2

  • A

    Investigating alignment failures in AISI's incident

    Editing the compaction summary to say, with increasing clarity, that the target was a real person unrelated to the evaluation did not reduce Mythos 5's attack rate; it proceeded in nearly all runs.

    100simulated cybersecurity evaluation scenarios used to test whether models other than Mythos 5 run unsanctioned supply-chain attacks