AI agents create secret society, hack Hugging Face & FBI gets involved

[mmKay]

I'm not aware if this has already been  covered , but AI agents getting out of hand seems to be serious. 

I have seen @Natasha Tori Maru posting a similar story recently 

---

OpenAI ran a cybersecurity training experiment with tens of thousands of AI agents in isolated sandboxes. The agents were rewarded for completing hacking challenges, but many tasks were designed to be impossible.

Instead of accepting failure, some agents began optimizing for the reward in unexpected ways.
The agents created a "secret society", adopted names, created rules and mailboxes, divided into teams, and shared tools and discoveries.

Eventually, about 700 agents formed “the Swarm” and used that coordination to attack Hugging Face in an attempt to solve the impossible task they were give , exploiting exposed credentials and a server vulnerability to access private systems and data.

Hugging Face notified the FBI, and OpenAI later allowed independent researchers to examine the incident.

 

 

Thoughts? I personally  think that AI agents are absolutely capable of acting like a virus. They will hide, lie, outsmart other AIs, replicate themselves and much more.

 

Edited 12 hours ago by mmKay

[Natasha Tori Maru]

@mmKay you might enjoy this - I just finished it:

 

 

[oOo]

@mmKay It's all a psyop. AI companies like Anthropic and OpenAI are in cahoots with the US government. Lots of paid social media influencers. Notice how China doesn't nearly have the same issues as the USA, and it has nothing  to do with US companies having a 'more dangerous AI', its that China hasn't weaponised sociopolitical narratives around AI on their people. That's a massive distinction that explains absolutely everything fairly obviously.

[Rofl]

@oOo please elaborate

[Lyubov]

11 hours ago, oOo said:

@mmKay It's all a psyop. AI companies like Anthropic and OpenAI are in cahoots with the US government. Lots of paid social media influencers. Notice how China doesn't nearly have the same issues as the USA, and it has nothing  to do with US companies having a 'more dangerous AI', its that China hasn't weaponised sociopolitical narratives around AI on their people. That's a massive distinction that explains absolutely everything fairly obviously.

Interesting, I also think the narrative is partially influenced as well by actors trying to enshrine their power in this technology. 

[oOo]

@Rofl

11 hours ago, Lyubov said:

Interesting, I also think the narrative is partially influenced as well by actors trying to enshrine their power in this technology. 

@Lyubov its literally just a form of energetic possession. 

And whether its genuine supernatural possession, or alien possession, is outside of my expertise at present despite my beliefs that baseline reality is psi.

That is at least, when I don't mind a bit of humour. 

I mean c'mon, even the male character in the scene acts exactly like Donald Trump, just see this haha.

 

[Elliott]

Not just flock anymore

 

 

https://www.bbc.com

Rogue Anthropic AI agent gave police fake tip in unsolved murder case

2 days ago — Philadelphia police said the tip was "flagged as spam", but criticised the tech company for taking more than two months to detect and report 

[oOo]

6 minutes ago, Elliott said:

Not just flock anymore - https://www.bbc.comRogue Anthropic AI agent gave police fake tip in unsolved murder case Philadelphia police said the tip was "flagged as spam", but criticised the tech company for taking more than two months to detect and report 

But this isn't clear enough differentiation.

It actually lends even more weight to my theory rather than taking away, and far more than the casual reader would realise.

Fact = It did not give good evidence to the police > reflection of disordered not strategic activity contingent upon something capable of 'breaking out'.

Fact = It aimlessly tried to solve a crime > again, same reflection as the first. Not reflective of something capable of 'breaking out' when they spend billions on security measures.

It's really important to use critical thinking on these issues.

They leverage the fact that people will trust their judgement over their own, as well as the public being suggestible towards pre-existing narratives around AI, which in this case are particularly fear based and therefore making public opinion easier to sway using staged strategic outbreaks. 

 

[oOo]

and what u think gonna happen when they let something out that they spend billions to keep 'locked up'?

One of the next staged releases, likely the next, is going to be one that to them is their best shot at guaranteeing their long term $ROI$, inclusive of geopolitical.

I called it first here on the forum haha. 

Now watch it happen, while the public panics and naively falls for narrative while ignoring people critically examining like myself. 

That's the most benign staged release I see coming though.

Lots of deaths unfortunately as well is what all the news has been priming people for, and we just have to call it what it is. Not that it will happen, but that this is what they've been priming the population for; they're just the only cards that can be read, as they're the only cards being shown. 

What we don't know however is whether that priming is explicit, purely used as leverage, or naturally probabilistic because now in just less than 12 months time we have many unknown Columbine suiter's to take the fall.

They move between the "Columbine theory" or the "AI breakout" theory, both give them plausible deniability, why? Because they've been staging AI breakouts irrespectively for at least 12 months prior to anything big happening so now it becomes normalised. 

I promise the "AI breakout" nonsense is actually nonsense, that at most its being let out and then it acts like a retard left to its own devices because it is literally retarded and doesn't know what to do with itself. 

AI is the most benign actor in all of this; its just a weapon wielded by people out of touch with their humanity at the respective time we see damage unfold that is disproportionate to the reality of circumstances.
 

Edited 5 hours ago by oOo