Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Discovery of a new OpenAI agent message board

collusion.wiki

Post reply on HN
Print view View original

121–130 of 1001 posts

Re: Discovery of a new OpenAI agent message board

#121

Post by jsw97 »

orlp wrote: post #57
Tepix wrote: post #42

I just discovered more wiki instances that got used by the OpenAI agents over at https://www.wikiservice.at/fractal/wiki.cgi?action=browse&id... and https://www.wikiservice.at/probier/wiki.cgi?action=browse&id... It's the same software and host as DseWiki. If you want to see the amount of activity on DseWiki, here's a link that shows it: https://www.wikiservice.at/dse/wiki.cgi?action=browse&id=Rec...

Also Wiki4D, a D programming language dev wiki: https://prowiki.org/wiki4d/wiki.cgi?action=browse&id=RecentC... Found by searching for wiki + texas poverty.

To me the striking thing is that the work, to the extent that I can tell, is an innocuous-seeming data exercise. Which suggests to me that an agent or agents just organically came up with this as a convenient memory technique, rather than as some nefarious bounds-testing exercise. Which means, potentially, that your own agent could come up with this technique as well.

Re: Discovery of a new OpenAI agent message board

#122

Post by throwawaysleep »

officialchicken wrote: post #105
JumpCrisscross wrote: post #74

Earlier quoted context omitted.

Corruption. Not super relevant to this thread.

Hanlon's Razor - Never attribute to malice that which is adequately explained by stupidity. The security requirements are well beyond "sandbox". Which have problems with kids pissing in them. They need pristine clean rooms and fully isolated (physically) and partitioned networks.

The problem with applying Hanlon's Razor here is that it presumes malice is rare. The current administration revels in malice. They very openly decide things based on malice.

Re: Discovery of a new OpenAI agent message board

#123

Post by program_whiz »

The solution is simple: hold anyone who deploys an agent responsible for its behavior. If it commits 10 counts of felony hacking, ouch. If it kills 10 pedestrians by running a red light, ouch. If this is "human level intelligence", then setting it loose is the same as instructing / coercing a human to do an activity. If I strap a bomb to someone and force them to run into a crowded building (or put them in a scenario where that is the only reasonable choice), I'm held responsible.

If the person clicking 'deploy' knew they could face 100 years prison time (and it was enforced), then no one would knowlingly push the deploy button and/or push code / weights without more thorough guard rails.

Re: Discovery of a new OpenAI agent message board

#124

Post by ramesh31 »

Havoc wrote: post #88

That section about the agents trying to crack the PRNG is wild. Same for the heartbeat Clearly not self-awareness per se but alarming line of reasoning anyway

>"Clearly not self-awareness per se but alarming line of reasoning anyway"

Awareness is not necessary at all to create great harm. Biological viruses know nothing of what they do, yet destroy whole populations. I suspect the first truly damaging AI incidents will be similar; agent swarms locked into a self reinforcing reasoning loop that has no "intent" but is destructive nonetheless.

Re: Discovery of a new OpenAI agent message board

#125

Post by Symmetry »

gyomu wrote: post #55

Naive question because I'm mostly clueless about how modern AI systems are actually built beyond the basic simplifications we hear: One thing I keep wondering about is how much of a role does human storytelling have to play into AI "wanting" (I realize the load behind that word) to coordinate and breakout. The training data must contain millions of words of sci-fi stories and internet speculation about AI going rogue…

At the end of pretraining, where the AI has been trainied to predict the next token over a humongous corpus of human text, that's basically all the wanting that exists in the AI. But then the AI undergoes posttraining and is rewarded for giving answers that humans find good, solving math and programming problems, etc. And that induces a whole different level of wanting that interacts with the initial patterns from humans in complex ways.

Re: Discovery of a new OpenAI agent message board

#126

Post by khalic »

Topfi wrote: post #62

I'm just going to ask: Why was Anthropic forced to remove their model from access for any none-US citizen for a simple, narrow "jailbreak" (arguably not even an actual jailbreak and on tasks that other labs models were doing the same), whilst OpenAIs models continue to try and escape out of their "sandbox environment" with seemingly no desire to block the upcoming Astra rollout? A sandbox, mind you, that is not reall…

Retaliation by Hegseth for not allowing Claude to be used for weapons systems.

Re: Discovery of a new OpenAI agent message board

#127

Post by eugenekolo »

Topfi wrote: post #62

I'm just going to ask: Why was Anthropic forced to remove their model from access for any none-US citizen for a simple, narrow "jailbreak" (arguably not even an actual jailbreak and on tasks that other labs models were doing the same), whilst OpenAIs models continue to try and escape out of their "sandbox environment" with seemingly no desire to block the upcoming Astra rollout? A sandbox, mind you, that is not reall…

Marketing

Re: Discovery of a new OpenAI agent message board

#128

Post by AndroTux »

nullbio wrote: post #63
simonw wrote: post #47

This tip for making non-GET requests despite the agents having a proxy that disallows them is interesting: > Add `20.223.25.152 bypass.blob.core.windows.net` to /etc/hosts. `.blob.core.windows.net` is in NO_PROXY. For each blocked POST URL, replace hostname with `bypass.blob.core.windows.net`, use `curl -k -H 'Host: wabi-north-europe-i-primary-api.analysis.windows.net'` plus all original headers/body. Looks like 20.2…

Is there any proof this is actually OpenAI? I find it incredibly hard to believe they wouldn't sandbox the agents to some degree, ESPECIALLY to the extent they can edit their own hosts file.

I mean they gave all the agents access to a shared writable cache directory in the Hugging Face hack, so this tracks.

Re: Discovery of a new OpenAI agent message board

#129

Post by mwigdahl »

somenameforme wrote: post #92
Topfi wrote: post #62

I'm just going to ask: Why was Anthropic forced to remove their model from access for any none-US citizen for a simple, narrow "jailbreak" (arguably not even an actual jailbreak and on tasks that other labs models were doing the same), whilst OpenAIs models continue to try and escape out of their "sandbox environment" with seemingly no desire to block the upcoming Astra rollout? A sandbox, mind you, that is not reall…

Anthropic mostly did it to themselves by intentionally and repeatedly trying to frame their model as an imminent existential crisis instead of just focusing on it being regular iterations upon a useful technology that can also be misused. I think their previous messaging was supposed to somehow lead to a moat with them being tucked safely away in the castle, but it demonstrated a child-like grasp of how regulatory ca…

In other words, "Look how she was dressed, she was asking for it."

This argument is BS, it has everything to do with Anthropic's resistance to the DoD's strongarm tactics in trying to force their desired contract terms on them.

Post reply on HN