Conversation

Notices

1. Embed this notice
   Paul Cantrell (inthehands@hachyderm.io)'s status on Friday, 31-Jul-2026 12:34:59 JST Paul Cantrell

   If you doubted that the “oh noes our agent escaped containment” schtick from OpenAI was anything other than marketing, note Anthropic’s “look look! we’ve got one too! oh no! look at us!!” story.

   • Embed this notice
     Paul Cantrell (inthehands@hachyderm.io)'s status on Friday, 31-Jul-2026 13:45:33 JST Paul Cantrell

     in reply to

     Both these replies point at a common thread: these news stories represent a grotesque failure of either the vendors or their products.

     If the agents did in fact “go rogue” and cause damage, even in the hands of the people who should be their most competent operators, then the product is unpredictable, unreliable, and unfit for purpose.

     If the product is •not• in fact unfit for purpose, then that means there's some combination of negligence and malice behind these incidents, and they are at •best• knowing harm in service of a marketing stunt or (as Parsons argue) something even worse.

     https://peoplemaking.games/@aubilenon/117012729959327465

     https://mstdn.social/@Npars01/117012715615344732

   • Embed this notice
     Paul Cantrell (inthehands@hachyderm.io)'s status on Friday, 31-Jul-2026 13:50:43 JST Paul Cantrell

     in reply to

     If these incidents were truly as represented, they'd be colossal embarrassments. The AI vendors would be trying to bury them, not trumpeting them.

   • Embed this notice
     Thomas Depierre (di4na@hachyderm.io)'s status on Friday, 31-Jul-2026 17:37:41 JST Thomas Depierre

     in reply to

     @inthehands here is my analysis.

     It is both. These posts are just... Delusional

   • Embed this notice
     Paul Cantrell (inthehands@hachyderm.io)'s status on Tuesday, 04-Aug-2026 01:37:34 JST Paul Cantrell

     in reply to

     I rarely link to videos that convey information that could have been a few paragraphs of written text, but this one is good enough that I’m linking to it. Lots of good thoughts in here that add to what I wrote upthread, expressed in a very accessible way.

     One worth nothing: the remarks at 4:30 about how the “alignment” framing obscures things we’d normally call “unreliable” or “broken.”

     https://sauropods.win/@futurebird/117032480870736138