I frequently find myself instructing Opus 5 to stop overcomplicating things.

It just wants to talk jargon heavy and add unnecessary noise to the conversation.

I wonder it’s related to text watermarking somehow.

latentsea 14 hours ago | parent | next [-]

I frequently find myself yelling out loud in frustration "speak fucking English Claude!" followed by asking to rewrite it ASD STE100 Simplified Technical English. It's gotten to the point where it's inducing burnout and I'm about to campaign for us to switch to Codex because of it. I just fucking can't anymore with Opus. It's beyond me why they did this to the model and aren't immediately revolted themselves enough to quickly back it out.

	▲	itopaloglu83 13 hours ago \| parent [-]
		Yes, maybe there should be an STE100 only mode. I still have some promotional credits and use them with Fable, and the answers are night and day in my particular use case.

▲

preg_match 8 hours ago | parent | prev [-]

I believe it is related to watermarking. Text can't encode additional information like a image can, so the watermarking has to come from the structure, word choice, voice, etc. Claude has to sound like Claude so Claude can tell if text comes from Claude.

I also think this is why the models aggressively drift back to the exact same word choice. You can put it in CLAUDE.md, explicitly prompt it, doesn't matter. A prompt or two later it's back on its bullshit.

	▲	itopaloglu83 7 hours ago \| parent [-]
		Exactly, and frequently it also responds to the latest prompt forgetting everything else in the conversation. I don’t remember where I saw it, but I think there’s also a smaller agent paraphrasing my requests to the agent because I remember being angry and saying aiming sarcastic about an issue and the thinking notes said the user is angry do this blah blah, but then the Claude claim I asked for that specific solution. I think they’re running some optimizations and multi level agents etc. tests to reduce costs.