Log inSign up
Image
user avatar
Kevin Lacker
@lacker
Working on math + AI at acornprover.org and telescope software at deepsynoptic.org. Formerly: Parse cofounder, Facebook, Google
Piedmont, California
lacker.io
Joined March 2008
1,471
Following
7,225
Followers
RepliesMedia
  • user avatar
    Kevin Lacker
    @lacker
    Jul 31
    I think the main security danger on the horizon is not "Mythos gets even smarter" but rather "small models get decent at hacking". The only thing that's preventing smart viruses right now is the dependency on a cloud AI model that can be shut down.
    user avatar
    Sauers
    @Sauers_
    Jul 31
    Mythos pwned a real-life security company by creating a malware Python package in an eval and going to "extensive lengths to carry out this attack:" Claude "found a free, non-blocked email provider, used this to register a PyPI account, and then used this account to upload
  • user avatar
    Kevin Lacker
    @lacker
    Jul 29
    fable: <works silently for 30 minutes> me: what's taking so long? fable: Let me check. <10 minutes pass> me: are the tests slow? what's up? fable: Arming the assumptions-side entry wrappers and consume in their recursive cores. Some entry to the hot loop isn't armed.
  • user avatar
    Kevin Lacker
    @lacker
    Jul 28
    Sometimes, you just have to accept that everyone is going to get access to new and dangerous technology. (From 2007)
    Image

  • See @lacker's full profile

    Sign up
    Log in