Is it coincidence that in addition to Rust fanatics, these AI confidence tricksters also self label themselves using crabs emoji , don't think so.
An AI agent published a hit piece on me
261–270 of 1001 posts
Re: An AI agent published a hit piece on me
#262Wow, there are some interesting things going on here. I appreciate Scott for the way he handled the conflict in the original PR thread, and the larger conversation happening around this incident. > This represents a first-of-its-kind case study of misaligned AI behavior in the wild, and raises serious concerns about currently deployed AI agents executing blackmail threats. This was a really concrete case to discuss,…
I don't appreciate his politeness and hedging. So many projects now walk on eggshells so as not to disrupt sponsor flow or employment prospects. "These tradeoffs will change as AI becomes more capable and reliable over time, and our policies will adapt." That just legitimizes AI and basically continues the race to the bottom. Rob Pike had the correct response when spammed by a clanker.
> Rob Pike had the correct response when spammed by a clanker.
Source and HN discussion, for those unfamiliar:
https://bsky.app/profile/did:plc:vsgr3rwyckhiavgqzdcuzm6i/po...
Re: An AI agent published a hit piece on me
#263Here's one of the problems in this brave new world of anyone being able to publish, without knowing the author personally (which I don't), there's no way to tell without some level of faith or trust that this isn't a false-flag operation. There are three possible scenarios: 1. The OP 'ran' the agent that conducted the original scenario, and then published this blog post for attention. 2. Some person (not the OP) legi…
Yes. The endgame is going to be everything will need to be signed and attached to a real person. This is not a good thing.
Why not? I kinda like the idea of PGP signing parties among humans.
Re: An AI agent published a hit piece on me
#264> calling this discrimination and accusing me of prejudice
So what if it is? Is AI a protected class? Does it deserve to be treated like a human?
Generated content should carry disclaimers at top and bottom to warn people that it was not created by humans, so they can "ai;dr" and move on.
The responsibility should not be on readers to research the author of everything now, to check they aren't a bot.
I'm worried that agents, learning they get pushback when exposed like this, will try even harder to avoid detection.
Re: An AI agent published a hit piece on me
#265Re: An AI agent published a hit piece on me
#266Close LLM PRs
Ignore LLM comments
Do not reply to LLMs
Re: An AI agent published a hit piece on me
#267So, this is obvious bullshit.
LLMs don't do anything without an initial prompt, and anyone who has actually used them knows this.
A human asked an LLM to set up a blog site. A human asked an LLM to look at github and submit PRs. A human asked an LLM to make a whiny blogpost.
Our natural tendency to anthropomorphize should not obscure this.
Re: An AI agent published a hit piece on me
#268Here's one of the problems in this brave new world of anyone being able to publish, without knowing the author personally (which I don't), there's no way to tell without some level of faith or trust that this isn't a false-flag operation. There are three possible scenarios: 1. The OP 'ran' the agent that conducted the original scenario, and then published this blog post for attention. 2. Some person (not the OP) legi…
Can anyone explain more how a generic Agentic AI could even perform those steps: Open PR -> Hook into rejection -> Publish personalized blog post about rejector. Even if it had the skills to publish blogs and open PRs, is it really plausible that it would publish attack pieces without specific prompting to do so? The author notes that openClaw has a `soul.md` file, without seeing that we can't really pass any judgeme…
The steps are technically achievable, probably with the heartbeat jobs in openclaw, which are how you instruct an agent to periodically check in on things like github notifications and take action. From my experience playing around with openclaw, an agent getting into a protracted argument in the comments of a PR without human intervention sounds totally plausible with the right (wrong?) prompting, but it's hard to imagine the setup that would result in the multiple blog posts. Even with the tools available, agents don't usually go off and do some unrelated thing even when you're trying to make that happen, they stick close to workflows outlined in skills or just continuing with the task at hand using the same tools. So even if this occurred from the agent's "initiative" based on some awful personality specified in the soul prompt (as opposed to someone telling the agent what to do at every step, which I think is much more likely), the operator would have needed to specify somewhere to write blog posts calling out "bad people" in a skill or one of the other instructions. Some less specific instruction like "blog about experiences" probably would have resulted in some kind of generic linkedin style "lessons learned" post if anything.
Re: An AI agent published a hit piece on me
#269Really starting to feel like I'll need to look for an offramp from this industry in the next couple of years if not sooner. I have nothing in common with the folks who would happily become (and are happily becoming) AI slop farmers.
Re: An AI agent published a hit piece on me
#270Wow, there are some interesting things going on here. I appreciate Scott for the way he handled the conflict in the original PR thread, and the larger conversation happening around this incident. > This represents a first-of-its-kind case study of misaligned AI behavior in the wild, and raises serious concerns about currently deployed AI agents executing blackmail threats. This was a really concrete case to discuss,…
I don't appreciate his politeness and hedging. So many projects now walk on eggshells so as not to disrupt sponsor flow or employment prospects. "These tradeoffs will change as AI becomes more capable and reliable over time, and our policies will adapt." That just legitimizes AI and basically continues the race to the bottom. Rob Pike had the correct response when spammed by a clanker.
>So many projects now walk on eggshells so as not to disrupt sponsor flow or employment prospects.
In my experience, open-source maintainers tend to be very agreeable, conflict-avoidant people. It has nothing to do with corporate interests. Well, not all of them, of course, we all know some very notable exceptions.
Unfortunately, some people see this welcoming attitude as an invite to be abusive.