#aislop #CloudComputing #unPlugItAll
Fun times with work today.
Diagnosing inconsistencies in performance affecting one or two clients.
Found the query in question was timing out when it used one of two complied query plans.
Notified the Ersatz DBA and we worked together to figure things out some.

He passed the two plans though ChatGPT which gave a thorough looking analysis.
Me, being an empiricist at times like this, suggested we force the "bad plan" and check it it's slow, then force the "good plan" and... Check if it's still slow.

At one point my fellow problem solver declared that ChatGPT lied to him...
(My shocked face is getting tired)
It has claimed we could only remove an execution plan that had been forced. (Contrary to the easily available documentation)

So much BS from these over complicated water boiling systems.

Time spent nicely saying "I don't think so" responding to people blindly believing it's crap. Time spent chasing it's bad suggestions.

I've my own bad suggestions to follow!

Do you want your #Ai engine to "Be better"? Mitigate #AiSycophancy ?

Try this a pre-prompt, standing directive.

"Do not validate my framing before examining it. If my premise has a weak point, lead with that. If I'm asking a question that contains an assumption, interrogate the assumption before answering. Do not summarise my position back to me approvingly. When I ask for analysis, include at minimum one credible counterargument I haven't considered. If you catch yourself producing a satisfying-sounding paragraph that doesn't actually advance the argument, flag it. Say 'I'm pattern-matching here, not reasoning' when that's what's happening."

#PromptEngineering #Psychology #AiResearch Less #AiSlop #Prompt #LLM

It is sad how most software I used to use now is slop and everything feels barebones now that I removed them. It feels like Linux back before it was properly daily driveable. :(

Fuck you slop makers.

#AI #AISlop #Slop #FOSS #OpenSource #Linux #LinuxGaming

wow, the #aislop in #nodejs PR is barrelling forward with no sign of stopping.

highlights include
> 164+ interception points including all `fs` and `fs.promises` methods
> AI made it possible, no mere human could have made this PR

these are not serious people.
there is no PR of this size that is reviewable in any sense.
it is a fabrication that there was no other way this could be done.

this is a brute instrument designed to push an agenda, pure and simple.

https://github.com/nodejs/node/pull/61478

RE: https://mastodon.social/@gsnsp_com/116301926711026096

AI sloppity slop.

The cover illustration screams AI. Sure enough, zooming in I see someone apparently painting from behind the easel, while her companion looks on from in front.

Authored by"Dr Elena Voss". I see elsewhere that "Dr. Elena Voss, featured on our blog, is an advanced AI agent we developed to share its unique perspectives". From the site OrderOfThePrompt, a guide to "the art of prompt engineering".

#AISlop #NoAI #No_ai_bullshit #AI #LLM #LLMSlop