Hacker Newsnew | past | comments | ask | show | jobs | submit | cs702's commentslogin

... in two hours:

> No credentials. No insider knowledge. And no human-in-the-loop. Just a domain name and a dream. ... Within 2 hours, the agent had full read and write access to the entire production database.

Having seen firsthand how insecure some enterprise systems are, I'm not exactly surprised. Decision makers at the top are focused first and foremost on corporate and personal exposure to liability, also known as CYA in corporate-speak. The nitty-gritty details of security are always left to people far down the corporate chain who are supposed to know what they're doing.


Linkbait headline. Without context, these figures mean nothing.

You can see the debt-to-GDP ratios here:

https://fred.stlouisfed.org/series/GFDEGDQ188S

https://fred.stlouisfed.org/series/GFDGDPA188S


Am I supposed to take away from these plots that we are all good since it's been steadily above the prior record in 1940 since 2020? Or is everything okay since it was really going up and it did course correct to a bit more of a straight line recently?

The article seems to be communicating that this rate of spending is not sustainable.


Japan debt/GDP is more than twice the US's.

https://economics.stackexchange.com/questions/11792/what-lev...

Not to say that it's okay, and Japan's economy certainly has issues with stagnation due to the debt load, but it's also not a "we have imminent hyperinflation" kind of thing either.

The concern with the past five months isn't so much the level of debt, it's the rate of change - we're increasing it faster than in the past... and this isn't a COVID-level crisis or a 2008-style deep recession either where Keynesian logic might make more sense.


> Japan debt/GDP is more than twice the US's.

Japan borrows on 0.75% interest rate compared to current US's 3.5%.



Is Japan still the largest creditor in the world???

The real question is what percentage of GDP is directly created (or continues to exist) because of the increased debt. When this metric was created the GDP was more authentic and not debt driven.

The article literary addresses this:

> Economists aren’t necessarily worried by the total level of debt (in fact, government debt is a necessary foundation of global markets). Rather it’s the debt-to-GDP ratio, which measures a nation’s borrowing against its growth



The article has more details than just the headline. For example:

> Maya MacGuineas, president of the Committee for a Responsible Federal Budget (CRFB), said that interest payments on the debt are expected to exceed $1 trillion this year, and will surpass $2 trillion by 2036.

That’s very concerning. There’s no plan to run balanced budgets and stop deficits. And no plan to reduce debt. And no plan on economic competitiveness against China. American politics is mostly dominated by irrelevant things that won’t fix the fundamental problems that will come to affect us in the future.


They can always at some point pull a modern day Nixon Shock and cancel everything. Make all that debt go poof.

The plan is a smash and grab for whoever is smart enough to scam there way through it amd then probably default like every other empire

That looks pretty bad, and getting worse.

TL;DR: The authors found current-generation AI agents are too unreliable, too untrustworthy, and too unsafe for real-world use.

Quoting from the abstract:

"We report an exploratory red-teaming study of autonomous language-model–powered agents deployed in a live laboratory environment with persistent memory, email accounts, Discord access, file systems, and shell execution. Over a two-week period, twenty AI researchers interacted with the agents under benign and adversarial conditions."

"Observed behaviors include unauthorized compliance with non-owners, disclosure of sensitive information, execution of destructive system-level actions, denial-of-service conditions, uncontrolled resource consumption, identity spoofing vulnerabilities, cross-agent propagation of unsafe practices, and partial system takeover."


> current-generation AI agents are too unreliable, too untrustworthy, and too unsafe for real-world use

...a completely unsurprising result, but it's nice to see published experiments.

Any agent system using current LLMs is likely to exhibit undesirable traits that derive from the training data.


> undesirable traits that derive from the training data

The research areas of model alignment and safety are attempting to address this fundamental problem - and have yet to solve it convincingly.

Problems like emergent misalignment can make things even worse.

https://www.nature.com/articles/s41586-025-09937-5


"The invisible hand" of free markets has become truly invisible...

The OP is exactly right, in my view: the current charade of paywalled-journal "peer review" is broken beyond repair.

See also from the same author:

* https://www.experimental-history.com/p/the-rise-and-fall-of-...

* https://www.experimental-history.com/p/the-dance-of-the-nake...

* https://www.experimental-history.com/p/lets-build-a-fleet-an...


Whether you agree with Hinton or not, this is a great video for general audiences.

Thank you for sharing it on HN!


It used to take years, decades, or centuries before a system could grow and evolve to be so complex and unwieldy, and so full of internal contradictions, that the whole thing becomes an incomprehensible tangle of hairballs. An example is the patchwork system of international, national, regional, and local laws we have at present, which has grown and evolved over centuries.

Now, it can take only a few days or weeks.



Thank you for sharing this on HN.

It's a worthwhile effort. If successful, Woxi can enable a large mass of scientists and engineers who don't have access to Mathematica to run legacy code written for it. Also, Woxi would give those scientists and engineers who regularly use Mathematica a non-proprietary, less restrictive alternative, which many of them would welcome.

How does Woxi compare to other "clean-room implementations"[a] of the same language?

--

[a] Please check with a lawyer to make sure you won't run into legal or copyright issues.


These lawmakers are not even wrong.

To be wrong, one must understand what one is talking about.

Sigh.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: