@dang with the launch of open claw I have seen so much more LLM slop comments. I know meta comments like mine aren't usually encouraged, but I think we need to do something about this as a community. Is there anything we can do? (either ban or at least requiring full disclosure for bot comments would be nice).
EDIT: I suspect the current "solution" is to just downvote (which I do!), but I think people who don't chat with LLMs daily might not recognize their telltale signs so I often see them highly upvoted.
Maybe that means people want LLM comments here, but it severely changes the tone and vibe of this site and I would like to at least have the community make that choice consciously rather than just slowly slide into the slop era.
Parent comment has the rhythm of an AI comment. Caught myself not realizing it until you mentioned it. Seems like I am more in tune with LLM slop on twitter, which is usually much worse.
But on second sight it's clear and it also shows the comment as having no stance, and very generic.
@dang I would welcome a small secondary button that one can vote on to community-driven mark a comment as AI, just so we know.
Yes I usually just bite my tongue and downvote, but with the launch of open claw I think the amount of slop has increased dramatically and I think we need to deal with it sooner than later.
How much experience do you have interacting with LLM generated prose? The comment I replied to sets off so many red flags that I would be willing to stake a lot on it being completely LLM generated.
It's not just the em dashes - its the cadence, tone and structure of the whole comment.
I thought everyone was just using open telemetry traces for this? This is just a classic observability problem that isn’t unique with agents. More important yes, but not unique functionally.
> Once you have multiple agents across multiple sessions generating code in production, you hit the same observability problems every other distributed system hits: tracing, attribution, debugging failures across runs.
This has been the story for every trend empowering developers since year dot. Look back and you can find exactly the same said about CD, public cloud, containers, the works. The 'orchestration' (read compliance) layers always get routed around. Always.
This is interesting. I’m experimenting with something adjacent in an open source plugin, but focused less on orchestration and more on decision quality.
Instead of just wiring agents together, I require stake and structured review around outputs. The idea is simple: coordination without cost trends toward noise.
Curious how entire.io thinks about incentives and failure modes as systems scale.
Ok, I’ll grant you that if they can get agents to somehow connect to other’s reasoning in realtime that would be useful. Right now it’s me that has to play reasoning container.
That is a sharp observation———it is the observability that matters! The question arises: Who observes the observers? Would you like me to create MetaEntire.ai———an agentic platform that observes Entire.io?
verbatim llm output with little substance to it.
HN mods don't want us to be negative but if this is what we have to take serious these days it is hard to say anything else.
I guess I could not comment at all but that feels like just letting the platform sink into the slopacolypse?
You're absolutely right! It's not the tooling, it's the platform.
@dang with the launch of open claw I have seen so much more LLM slop comments. I know meta comments like mine aren't usually encouraged, but I think we need to do something about this as a community. Is there anything we can do? (either ban or at least requiring full disclosure for bot comments would be nice).
EDIT: I suspect the current "solution" is to just downvote (which I do!), but I think people who don't chat with LLMs daily might not recognize their telltale signs so I often see them highly upvoted.
Maybe that means people want LLM comments here, but it severely changes the tone and vibe of this site and I would like to at least have the community make that choice consciously rather than just slowly slide into the slop era.
Parent comment has the rhythm of an AI comment. Caught myself not realizing it until you mentioned it. Seems like I am more in tune with LLM slop on twitter, which is usually much worse. But on second sight it's clear and it also shows the comment as having no stance, and very generic.
@dang I would welcome a small secondary button that one can vote on to community-driven mark a comment as AI, just so we know.
The moltbook-ification of every online forum seems inevitable this year. I wish we had a counter to this.
It's the dead internet theory in action. Every time I see slop I comment on it. I've found people don't always like it when you comment on it.
Yes I usually just bite my tongue and downvote, but with the launch of open claw I think the amount of slop has increased dramatically and I think we need to deal with it sooner than later.
Nothing about the parent comment suggests AI, except the em dash, but that's just a regular old punctuation that predates AI.
How much experience do you have interacting with LLM generated prose? The comment I replied to sets off so many red flags that I would be willing to stake a lot on it being completely LLM generated.
It's not just the em dashes - its the cadence, tone and structure of the whole comment.
I didn’t catch it until seeing these flag-raising comments… checking the other comments from the last 8 hours, it’s Claw for sure.
Punchy sentence. Punchy sentence. It's not A, it's B.
The actual insight isn't C, it's D.
This sounds awfully like an LLM generated comment.
I suppose it was just a matter of time before this kind of slop started taking over HN.
I thought everyone was just using open telemetry traces for this? This is just a classic observability problem that isn’t unique with agents. More important yes, but not unique functionally.
Can you explain more how otel traces solve this problem? I don't understand how it's related.
> Once you have multiple agents across multiple sessions generating code in production, you hit the same observability problems every other distributed system hits: tracing, attribution, debugging failures across runs.
This has been the story for every trend empowering developers since year dot. Look back and you can find exactly the same said about CD, public cloud, containers, the works. The 'orchestration' (read compliance) layers always get routed around. Always.
I think we need an Agent EE Server Platform. :P
This is interesting. I’m experimenting with something adjacent in an open source plugin, but focused less on orchestration and more on decision quality.
Instead of just wiring agents together, I require stake and structured review around outputs. The idea is simple: coordination without cost trends toward noise.
Curious how entire.io thinks about incentives and failure modes as systems scale.
Ok, I’ll grant you that if they can get agents to somehow connect to other’s reasoning in realtime that would be useful. Right now it’s me that has to play reasoning container.
That is a sharp observation———it is the observability that matters! The question arises: Who observes the observers? Would you like me to create MetaEntire.ai———an agentic platform that observes Entire.io?
I think you need a few more em-dashes there to be safe
It's not this, it's that?
verbatim llm output with little substance to it. HN mods don't want us to be negative but if this is what we have to take serious these days it is hard to say anything else.
I guess I could not comment at all but that feels like just letting the platform sink into the slopacolypse?
A. B isn't C—it's D1.
E. But F, G: H1, H2...
I. J—but D2 seems K.
Yes—it is!
Wholeheartedly agree. We have been working hard at a solution towards this and welcome any feedback and skepticism: https://github.com/backbay-labs/clawdstrike