Elektrine lite

← Feed

@norootcause@hachyderm.io

2026-08-23 04:55 UTC

I wrote up some thoughts on Boris Tane's argument for putting AI agents on-call. https://surfingcomplexity.blog/2026/08/22/wild-ai-related-reliability-incidents-are-coming/

Replies (3)

  • @jmeowmeow@hachyderm.io 2026-08-23 05:32

    @norootcause@hachyderm.io I had hoped that Theory of Constraints would catch us before Ironies of Automation, but you correctly identify automated on-call incident remediation as an area fully exposed to the Bainbridge Syndrome: when things go bad in an automatic controls system, they go horribly wrong.

    Open ##4539885

  • @bearbob@hachyderm.io 2026-08-23 05:21

    @norootcause@hachyderm.io I think your outlook makes a lot of sense. The premise of “if the agent is stuck it should page a human” also assumes the agent will somehow notice it is stuck. But given all I have seen so far in LLM Operators, they will not see a problem and take a step back, but happily jump to solve it. Even if, as you wrote, it doesn’t make sense or actually turns out to be harmful (“You are absolutely right, the core issue is different from what I first suspected”)

    Open ##5173547

  • @Di4na@hachyderm.io 2026-08-24 06:42

    @norootcause@hachyderm.io re using zero day being unreasonable. I did it in the past, I have seen multiple people do it, I do not consider it unreasonable. You may need to qualify or extend on it, because that is to me a perfectly reasonable and acceptable solution to Get Stuff Done.

    Open ##5173567