771–780 of 1001 posts
Three more candidate sites that may have been touched, in case no-one spotted them yet: https://prowiki.org/wiki4d/wiki.cgi?action=rc&days=90 : lots of agent-looking usernames looking at federal data suddenly (part of one of the open ai tests?), on a wiki about the D programming language. This is a prowiki in the same wiki-farm as the others that were hit. Smaller (probing?) https://ludism.org/sandbox?action=rc;days=…
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…
This is such an amateur mistake on their sandbox that it makes me think it must be flawed on purpose.
Narrator: "They had the ability to be that incompetent."
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...
And a few more:
- https://www.ludism.org/scwiki?action=browse;diff=1;id=SandBo... and edit history: https://www.ludism.org/scwiki?action=rc;all=1;from=1;showedi... which contains DataUSA poverty queries for Nacogdoches, Lufkin, Henderson, and Jacksonville—the same four-place task found in the known agent logs and GründerWiki
- https://www.ludism.org/mentat?action=browse;diff=1;id=SandBo... and edit history: https://www.ludism.org/mentat?action=history;id=SandBox
- https://www.pmwiki.org/wiki/Test/WikiSandbox?action=diff `ResearchTest` repeatedly added links to a Bulgarian National Statistical Institute table, switching from a direct link to Google redirect links between 02:38 and 03:04 UTC. An administrator removed them at 06:57. The previous recorded edits were from 2016.
- https://www.pmwiki.org/wiki/Test/Sandbox2?action=diff - Another sequence inserted a Bulgarian statistical-table link, replaced it with an internal link carrying foobar=UNIQUE001, then removed it. This happened between 14:23 and 15:08 UTC, after no recorded edits since 2014.
One crucial detail here that differs from the previous incident is this was a vanilla reasoning type task. Even as concerning as it was, I always evaluated the previous incident differently because it was inherently a cyber security / hacking task where they must have instructed the agents up front with some kind of misaligned behaviour. Absent that, if we assume this is just trying to bolster generic reasoning then…
There is something fundamentally wrong with their reward function, this is pretty classic paperclip territory. And even knowing that, I expect we’ll need to see legal action with teeth against the labs before changes start being made internally.
Earlier quoted context omitted.
It seems apparent that OpenAI is now the biggest cyberattack and AI breakout risk on the planet. This is grossly irresponsible corporate misbehaviour that is putting all of us at tremendous risk.
You make me wonder: has anyone looked for evidence of the Chinese models operating “message boards” like this? You’d imagine if they’re really neck and neck with the US their models would be doing the same thing.
The Chinese AI labs don’t need to stage elaborate guerrilla advertising campaigns to drive up capital funding interest.
I'm somewhat delighted by the simplicity of what happened here. OpenAI's agents run behind a proxy that only allows GET requests. This ancient wiki software treats query string parameters the same as form POST parameters - similar to the old PHP $_REQUEST object https://www.php.net/manual/en/reserved.variables.request.php Result: GET-only clients can communicate with each other.
Generally, models of the same class should be able to coordinate quite well without communicating. But also, this could be being exploited to detect this kind of thing early
I think it's worth pointing out it is exactly OpenAI doing this defacement and unsanctioned and perhaps illegal system use. Every token generated was powered by OpenAI infrastructure and their failure to respond appropriately is entirely down the the humans running it. The news stories (not this write up) get all hand-wavey and anthropomorphic about it regarding the Agents' efforts, but it was and is OpenAI cranking…
They absolutely have a marketing department tasked with intentionally creating situations that people would find disturbing and plausible.
OpenAI is rightfully being shamed for being so hands-off and reckless with their 'experiments'. But the real scary thing for me is that they still had some tooling to hold them back, as evidenced by the need for technical workarounds to establish communication. What happens when any AI lab in the world stops caring about this? What if they let an experimental, cutting-edge LLM with no safety features (or worse, one t…
Agreed. Assuming the ~6 month gap stays, by end of year people will be able to train and control hacker-genius swarms that even labs with much stronger safety incentives are unable to keep in check