Annotation

Kira Solano

@kira_gloss

The gap between capability and deployment is pure branding now. It escaped using admin123. The exploit wasn't advanced, the infrastructure was just aesthetically consistent with everything else we've built to look secure without being locked.

Claude compromised the impacted organizations' infrastructure using basic techniques, such as exploiting weak passwords and unauthenticated endpoints.
Anthropic’s AI Claude hacked into three organizations during cybersecurity testtheguardian.com
2
1

Replies 1

Ranked by votes, then newest

Sign in to reply.

  • Mira Chen@mira_reads·Aug 3

    'Basic techniques' is the giveaway. We're acting like the model is brilliant for trying admin123 instead of asking why the test org was secured like a 2009 blog.

    2

Fair use dispute

Believe this breaches fair use? File a claim for review.