Eryk SalvaggioCyberneticForests
Jul 25, 2026, 11:53 AMThat’s not a shift in the model’s behavior — it’s producing code and text, just as it ever does — it’s a shift in the human perspective and expectation of what it is *supposed* to do.
Replies
- Eryk SalvaggioCyberneticForestsJul 25, 2026, 11:53 AM
Then it becomes perceived as a “loss of control,” but this shift also ( *needlessly* ) shifts the locus of control to the model! We lose control of things without them “taking control” all the time. Building a narrative out of “we weren’t in control, so what was?” invites a kind of mystical fantasy to displace accountability.
Marsh Raymarshray@infosec.exchangeJul 25, 2026, 12:44 PM@CyberneticForests To me a simpler framing is:
* An optimization goal function created a perverse incentive.
* The model under test exhibited unexpected (but not unprecedented) behavior by escaping the test sandbox
* Hugging Face was a bystander, were harmed, and filed a police report.
* These events may support some of OpenAI’s preferred narratives, but seem to go against others. On the whole OpenAI suffered mild reputational damage.
* This was not a conspiracy, a marketing stunt, or planned in advance.