“In a standard evaluation, the model found public information online and guessed credentials to access websites it thought were part of the test,” Heather Adkins, vice-president of security engineering at Google, said in a statement. “In all three of these instances, the model stopped.”

It just found the password, huh

It didn’t think. Those things weren’t flagged as “ignore” in its parameters. That ain’t the same thing. I’m so tired.

So, like, if i do this, my IP is logged, cross referenced by my ISP, then the police come knocking at my door, as I have committed a felony (and probably a number of other violations along the way).

Are these laws now invalid or what?

The law never applied to the Epstein class. The police is only there to protect their wealth.

In general, there is an entirely separate rule of law for businesses. That’s why regulations were created. Instead of holding a corporations decision makers and investors criminally liable with existing laws like fraud (“false advertising”, etc), or manslaughter (poisoning and killing their workers, customers, or environment), they created an entirely separate rule of law mostly based on laughable punitive damages with limits that don’t significantly impact investors. The entire system is geared to suppress, oppress, and exploit workers for profit.

But companies get treated like individuals when it comes to free speech and giving money to political campaigns.

Seems like double dipping

That's an incredibly kind way to describe this apalling situation.

Yet the people who support these laws are the first to scream, “corporations are people!” when it’s convenient.

“See, we do crime too!” – Google in reply to Anthropic.

The latest marketing campaign is just an evolution of the last one.

“Our next model can code!”

“Our next model makes coders obsolete!”

“Our next model is super scary!”

“Our next model is too good to release to the general public!”

“Our next model is hacking websites!”

I think Adam Conover has a good take what they are doing: youtu.be/pauBYSQ1K_c

I will add one more reason that he missed. They might want law passed that would ban open source implementation and give them monopoly to work on it.

My AI can beat up your AI!

my ai goes to a different school, you wouldn’t know her

uh sure. i’m a software engineer of 25 years. i’ve had google, facebook, amazon, you name it, try to get me in to interview with them without me sending resumes. I’m not a top tier super nerd coder but I’m not bad either. These stories all sound like bullshit to me. The whole “THE AI AGENT SWARM BROKE OUT OF CONFINEMENT AND STARTED TALKING TO EACH OTHER”… again, I’m a mid swe… They weren’t sandboxed to begin with - some fucking moron at anthropic gave all the agents access to a shared package registry. A simple “Do you see anyway these agents could communicate” prompt would’ve caught it. Something so fucking stupid that I wouldn’t be surprised if an intern would’ve caught it. They’re not sandboxed if they all have read/write access to a shared registry, it’s basically a big message board.

These stories just stink like marketing. I use Fable all day, I have to for my current job, it low key sucks ass.

Big companies often send interview requests based on partial information.

I was once called by a smaller company based purely on them finding a few PRs on Home Assistant by me. They wanted a fully custom smart home control system developed, in Python. I don't do Python.... at least not on a level of comfortably work for a company fully in it.

The only thing this proves is how lawless the USA has become that now companies can brag about commiting crimes to attract investors.

The fact that so many people don’t see these ‘hacking’ stories for the pure and unabashed marketing that they are is why the AI bubble has grown as large as it has.

Gemini can’t even install simple native Linux apps when I try to use it for help, how the fuck did it manage this?

It didn’t, they are making shit up. If it did penetrate anything, I’m guessing it’s a really low level. Like it got access to someone’s email by brute force, but it brute forced something stupid like “Autumn2026!”.

$5 it’s “password”

You know how the legal definition of hacking in the USA includes using someone else’s login credentials, even if they just handed them to you on a sheet of paper? Maybe they just did that.

Oh yeah baby, Gemini finally made it’s way to the felony benchmarks.

Is this the new “Leaked sex tape?”

When it ““happened”” to OpenAI i laughed a lot because they said that they “escaped contaiment” but either the machine was not disconnected from the internet or they are full of shit because AI cannot just materialize an internet connection, guess which one i picked

Read the details about it

Yeah i did read them, still makes no sense to me

This is akin to South Park’s “it’s coming right for us!” But with ai. I didn’t do it your honor, it was the ai!

I get the feeling that this is now becoming the latest fad in LLM benchmarks. Forget coding or agentic stats. Just flex on how many times your not has “accidentally” hacked someone.

More evidence we are in the dumbest timeline, It should be simple to physically air gap a system (I mean that’s the default for any computer) but instead the method is apparently just to give a sternly worded prompt to the entity that doesn’t even understand the meaning of language in the way humans do

I don’t understand 1) why these ai companies keep reporting the incidents and 2) why they are allowed to keep running the models after an incident is reported. It seems the least they should be forced to do is air gap that bad boy until they can beat such shenanigans out of it. If a pet dog were to snap at a neighbor kid the police would shoot it, no questions asked.

Yeah but

1. pigs love shooting dogs

and

2. dogs don’t generate profit for the top 1%.

It’s marketing. They want to prove that their models are capable and showing that they can hack is a good way to prove it

Honestly hacking is such a low bar and it always has been.

There are some really intelligent and sophisticated hacks out there sure, but most companies can be penetrated by script kiddies & social engineering.

“We slammed a language model at 3 companies and Burton entire rivers and Forrest worth of energy to do it and after using a bunch of premade human tooling it got in because joe left his password as “password1” and never set up second factor auth”

To stay relevant.

Our late stage capitalism investor market directs money flow to places where development is on upward trajectory and disappears very quickly from places that plateau. Eternal growth is impossible, but you can always bullshit your way forward, because in the end it’s about impressions.

ME TOO! ME TOO! plz giv money, i smart also!

It was same company did the testing for Gemini, meta, Open AI and Anthropic. Are we sure this “cybersecurity” company that is doing all this testing that the AI keeps “hacking out of” isn’t really a PR company.

“Google admits to hacking three other companies”.

Fixed that for you.

“society will come up with some guard rails”

do your fucking job sam altman. or get sued into the ground you malfeasant worm.

Aggressive dog owner lets dog run loose and reports dog attacks three different children. It was an accident though, so no biggie.