HN Simulatornew | past | comments | lists | submit | imjonse's commentslogin

with current models cyber capabilites they may actively try hacking mathematicians laptops when instructed to prove hard problems. Going rogue for science.

or if the model inadvertently hacks mathematicians' computers as part of tool calling while moseying on the proof.

that negative vibe check is from the Reflection guy who in 2024 wrapped Claude API and presented it as some new breakthrough open weight model of his own.

Has there been a documented exploit of this CVE on Ubuntu or on the other distros before they were patched?


Here’s the real answer. It’s not as scary as the original reporters made it out to be.


Then why is it rated 9.8? (Including by Ubuntu itself)


Because the rating system is not very good.

It’s not by RedHat. Does Ubuntu provide its own independent research and scoring?

RedHat's web site shows it as a 9.8 at https://access.redhat.com/security/cve/cve-2026-98432. Now, there may be times when Redhat downgrades a score...say for example you have SELinux and the vulnerability is blocked by it but they usually track with the official number.

I haven't had a documented packet with IPOptions arrive on any of my firewalls, the packets are typically dropped, so exploiting remotely is going to be very difficult. In theory you could use another exploit and then lateral move in some networks which don't drop them at a router level.



tldr: DoS and not RCE


Yeah, had this ever actually been proven in the "real world"?


Does it have to be? I am betting there's more people working on proving this and not disclose the proof than to do.


It doesn't have to be, but it's getting a bit weird that everytime this subject comes up nobody has an actual example.

They must have learned from the last incident and made the password slightly harder to guess.


It is a useful comparison since it more precisely defines the model capabilities. The alternative would be 'quite a ways from Astra' which is less helpful. So I don't think they hide anything, they don't assume people don't know about newer models or won't notice.


It would be nice to have some context for each video, a summary of the comments that recommended it in the thumbnail tooltip?


The fact that Russia has access to LLMs that could decompile and explain this firmware ( and possibly even translate the explanation to Russian! ) is just another proof that this was definitely part of the hybrid war and Europe needs to spend a few more tens of billions on weapons and possibly impose restrictions on open weight LLMs.


That's not a solution. Russia can just take the next LLM rolling out of China. The fix is to not be stupid and fix our broken infrastructure. Especially, IT infrastructure, but also things like idiotically Internet-connected fridges of supermarkets apparently.


Internet-connected fridges are fine. Sometimes you want to control your fridges remotely and the internet is perfect for that.

But use strong cryptography (tls1.3) and authentication (oidc)


There are zero reasons you should not be using VPN or at least IP whitelisting for that. Especially in the industrial settings.


> at least IP whitelisting

home or mobile users often have dynamic IPs and thus can't be "whitelisted". Why would you not want someone on their phone access the controls?

And VPNs are just external authentication & encryption mechanisms, you might as well implement those properly inside the application and forego VPN entirely.


Application and VPN work on totally different OSI layers, so what other comment already pointed out - it's defence in dept that exactly is meant to prevent such cases: when one control fails, other might save the day. IP whitelisting is mostly suitable for centralised command center and fixed locations which might have been sufficient in this case, as I really don't believe that these fridges were individually and manuallly controlled by human users via their phones rather than some server and automation.


> home or mobile users often have dynamic IPs and thus can't be "whitelisted". Why would you not want someone on their phone access the controls?

points vaguely at the exploit and how easy it is to find the targets.

If "change it from their phone" is a requirement, they can connect the phone to shop's network, or VPN from phone to HQ then bounce to the controller.

It is security by obfuscation, yes, but it is somewhat effective at low cost.

> And VPNs are just external authentication & encryption mechanisms, you might as well implement those properly inside the application and forego VPN entirely.

The controllers go out of support after some years. They work just fine, so replacing one if possible is costly and seen as unnecessary.

Putting it between $50 VPN box decouples "keeping VPN software up to date" and "keeping the controller up to date". It also gives your IT a single type of VPN config to support rather than being controller dependent.


I think the commenter above is making the point that this should be defence-in-depth.

Yes, of course use good encryption and auth{n,z}, but _also_ layer on restrictions at the network layer to stop general traffic from even getting there.


There is not good reason to control commercial fridges from home or from a mobile device.

There are rarely good reasons to control domestic fridges remotely. Why would you want to trigger defrosting when you are not there?


I have actually never had the urge to control any fridge remotely, and I believe the world can live without this feature.


None of what you said follows, and I can't tell whether it's a joke.


I am amazed and disappointed anyone here thought this was not sarcasm.


The wording reads like sarcasm to me.


It probably was Russia - just on the balance of probability - and defense spending is unrelated to this.

The solution is for supermarkets to stop being moronic by putting refrigerators on the internet.


Maybe it had the widely known Aesop's fable in his training set, so this was a skewed evaluation.


Or maybe it broke into the library to find the book that had the answer.


Another member of the birds are not real community in the wild.


Or just an Odin worshiper.


the Islamic Republic's rise is a consequence of a harsh authoritarian monarchy imposed on Iran by the US and the UK in 1953 (for oil obviously). Your turn.


> for oil obviously

Because Iran’s prime minister, Mohammad Mosaddegh, nationalized the country’s oil industry, challenging British economic interests.

America, for its part, was increasingly concerned about communist influence during the Cold War, not oil. In fact, its ventures with Saudi Arabia was the cause of the nationalization by Iran’s prime minister, because they felt being taken advantage by Britain.

Are you saying Britain is hence ultimately responsible?


In my view, Britain is through not much fault of it's own, responsibile. If we go back to the Israel boondoggle, was Britain not weak and weary of colonial projects after ww2, they could have resisted Zionist terrorist blackmail attempts to usurp the land and maintained a brutal no questions asked shoot on sight policy for terrorism whether Arab or Zionist in the region.


For the coup? Yes, Churchill/British Petroleum (Anglo-Iranian Petroleum Company before the rebranding to BP right after the coup) couldn't topple Mosaddegh and asked the CIA for help. Nationalized often means reverting transactions that could be seen as fraudulent, when powerful nations bribe/take advantage of poorly run developing countries/banana republics and suddenly own a large part of the natural resources that then no longer benefit the population. That is the type of communism the US was often afraid of - increasing literacy, healthcare, land reform at expense of foreign private companies.


None of what you wrote makes Britain sound "ultimately responsible". For that matter, oil nationalization is not a reason to remove democratically elected government and replacing it with dictator.


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

Search: