by solenoid0937
4 subcomments
- "Sonnet sees the same two “obvious” bugs. It just cannot close the exploitation step. Mythos’s entire frontier advantage over the prior model is therefore bupkis."
What a bizarre conclusion. It "just" cannot close the exploitation step? "Just?"
Developing the working exploit is the hardest part, not finding the bugs. A self-proclaimed security professional should know this.
How is this stuff even making it to the top of HN? Is it just the trendy Anthropic hate? I wonder if these folks will publicly walk back their statements if Mythos turns out to be legit.
by lubujackson
0 subcomment
- Apparently this doom marketing strategy is working for landing enterprise deals, but boy these AI companies are stirring up consumer hate and fear.
I think the real purpose of the Mythos security sham is to mask that Anthropic simply can't release their new model because their data centers are already on fire. There are so many other red flags pointing to this: the no-Claude-Code-for-Pro-users "test", the AWS data center rental deal, the fact Microsoft rug pulled hard on Copilot, specifically removing Opus... and that's just the past 2 days?
- I wouldn’t be surprised if the glasswing thing comes with an NDA akin to what the NSA wants you to sign when you join. That would be the Anthropic-optimistic interpretation of the sound of crickets from participants - and ‘responsible disclosure’ would be an ok-ish reason for Anthropic itself to not publish what they found themselves alone.
If it’s indeed as bad as the article says it’s going to be a (yet another) PR disaster, but it won’t matter one bit as the whole industry is compute-constrained, not reputation-constrained. You’ll shout at clouds and them and their competitors and still be paying for tokens.
- A cunning move from Anthropic, nevertheless - their system has now seen and trained on a many big organization's source code.
- Am I supposed to know what a “system card” is?
- > The 244-page technical artifact, the thing that would have to survive peer review, refuses to actually quantify.
In what world does this author live where the system card is meant to be a scientific paper?
It's worth being skeptical, but it's nonsense to assume that the system card is meant for him or anyone to be able to reproduce and determine what the model actually did or did not. We won't know that until it is actually available.
- off topic: may I ask why and how this has been flagged? How does flagging work and why do I still see the post on the second or third page listed?
by yadhavkn15
0 subcomment
- website looks like fitgirl
- How about the boy who called nonsense security vulnerabilities. This is the same author who posts with incredulity that the ability to change a config file with a shell command in it gives you the ability to run the shell command you posted and wants it treated as some big CVE. Absolutely inconceivable that you might already have your harness in a sandbox where this is okay, and inconceivable that anyone might have a threat model that says that someone who can edit configuration of a tool can make that tool do arbitrary things allowed by its config.
https://www.flyingpenguin.com/ox-security-report-anthropic-m...
Hacker News