ivn

Depends on what you need a VPN for.

I don’t know them but they seems to be from Estonia marketing themselves for evading censorship. Are you in a country that censor VPN? If not why do you think the tradeoff might be worth? The 2 hops should have an impact on bandwidth and latency and there is probably no port forwarding.

Using a VPN is just moving your trust from your ISP to the VPN. Why would you trust a VPN you know nothing about giving them access to all your internet traffic?

That’s not my conclusion, that’s just mostly coming from the Guardian article. I say mostly because you’re missing one part, we know how the LLM is used.

That’s why I’m asking you to source your “conclusion”.

It wouldn’t change anything and I’m confused as to why you think it would and why you think I’m “giving a huge amount of benefit of doubt”.

I’m just pointing at what we know, what we don’t know and what you are just making up.

There is nothing that indicates that Anthropic’s AI is used to analyze data, I’m not saying it’s not, just that we don’t know. I’m going to quote a smaller section of a quote I made earlier of the same Guardian article:

In late 2024, years after the core system was operational, Palantir added an LLM layer – this is where Claude sits – that lets analysts search and summarise intelligence reports in plain English.

But the term AI is an issue here, there are multiple, of different kind, made by different companies. There is AI used for targeting, no doubt, but it’s not Claude, it’s Maven and some other subcomponents. The fact that Anthropic joined the project late, after it was already operational, is a good hint that they do not bring a core feature, but that’s only speculation.

I’m just saying that, as far as we know, the Anthropic contract is about Claude and the targeting is not made by a LLM.

They only mention Claude, where is the source that “some custom AI system made by Anthropic”, not a LLM, “was in the kill chain”?

I mean, I get that you want to tie Anthropic to this, I don’t like them either but we should stay factual and avoid filling the gaps with some “probably”. It’s also counterproductive as Maven and Palantir are huge menaces and this shift the blame away from them.

Well you’ll need to source your claim. The wiki article you linked only mention Claude.

The Anthropic contract is also quite recent compared to Maven creation.

I suggest you read the article.

The AI underneath the interface is not a language model, or at least the AI that counts is not. The core technologies are the same basic systems that recognise your cat in a photo library or let a self-driving car combine its camera, radar and lidar into a single picture of the road, applied here to drone footage, radar and satellite imagery of military targets. They predate large language models by years. Neither Claude nor any other LLMs detects targets, processes radar, fuses sensor data or pairs weapons to targets. LLMs are late additions to Palantir’s ecosystem. In late 2024, years after the core system was operational, Palantir added an LLM layer – this is where Claude sits – that lets analysts search and summarise intelligence reports in plain English. But the language model was never what mattered about this system.

Yes, but not for targeting, as explained in the article I linked.

The Maven Smart System is the platform that came out of those exercises, and it, not Claude, is what is being used to produce “target packages” in Iran.

Maven is doing the targeting, not Claude.

https://www.theguardian.com/news/2026/mar/26/ai-got-the-blame-for-the-iran-school-bombing-the-truth-is-far-more-worrying

Also the WaPo is now Bezos’ news.

You’ll need to read the documentation to understand its concepts first. You don’t use mkfs but zpool to create a pool then use zfs to create the fs on it. ZFS is usually used with some sort of raid configuration so most doc will show this but you can create a pool with no replication using a single drive.

For ZFS check the instruction for your distribution here: https://openzfs.github.io/openzfs-docs/Getting Started/index.html

NixOS so I can keep my config in git. I have a single nix config for all my machines (desktop, laptop and server) so I can share configuration between them. I use it to configure both my system and my user config, my dotfiles, with home-manager. Even my neovim config is in nix thanks to nixvim.

I don’t think I could go back now. It can be a bit of a pain from time to time and the learning curve is steep but it has so many advantages. Being able to rollback between config versions (called generations), having a consistent config between my machines, having it all in version control… The repo have so many packages and when there is a module it’s really easy to add a service. Writing new packages (derivations) and modules is also not that hard. It can be as simple as calling nix-init.

Had my main ssd fail on me a few month back and it was very simple to just replay the config and just get everything working as before. I only had to do the partitioning by hand (it can be done by nix but I’ve not gotten around to it yet). That’s why I only backup data and home partitions, not system partitions.

I highly doubt the users could get away with such an excuse but I’m not a lawyer. And the complaint comes from the football ligue rights holders, there is not ethics to be found here.

selling piracy services

That’s exactly the subject here.

France copyright law does not allow full length reproduction, whatever the intent. It does not apply here and it’s not why they are going after the customers here.

Telegram never was the private alternative to anything, unless you took their advertisement at face value. It was always known to not use e2ee by default.

Signal is not a problem, the US gov things were very dumb users issues. It was not caused by Signal itself.

So the goalpost has not moved in years, Signal was and is still good, Telegram is only fine if you do not care about privacy.

That’s not my claim.

Lol, “do your own research”.

This is not about me but about the advice you give to people. If you give advices you need to back them up, you can’t just answer “I disagree” to contradiction.