If the safeguard against mass surveillance is strictly tied to geolocation (US vs. non-US), it can't be an intrinsic property of the model. It has to be enforced at the API or contractual level. This means international users are left out of those core, embedded protections. Unless Anthropic is planning to deploy multiple, differently-aligned foundation models based on customer geography or industry, the safety harness isn't really in the model anymore.

Credit where it's due, going on record like this isn't easy, particularly when facing pressure from a major government client. Still, the two limits Anthropic is defending deserve a closer look.

On surveillance: the carve-out only protects people inside the US. Speaking as someone based in Europe, that's a detail that doesn't go unnoticed. On autonomous weapons: realistically, current AI systems aren't anywhere near capable enough to run one independently. So that particular line in the sand isn't really costing them much.

What I find more candid is actually the revised RSP. It draws a clearer picture of where Anthropic's oversight genuinely holds and where it starts to break down as they race to stay at the cutting edge. The core tension, trying to be simultaneously the most powerful and the most principled player in the room, doesn't have a neat resolution.

This statement doesn't offer one either. But engaging with the question openly, even without all the answers, beats silence and gives the rest of us something real to push back on.

Either "we" create models better than commercial state of the art (by using whatever means).

Or we use open models AND fund organisations building such models (could be by purchasing service from these orgs or donations - in which case would these orgs be different than hyperscalers?).

But i dont see how just hosting the models on some private servers would give us an edge?

You can give it a try: there are very easy cameras!

For matters concerning IT and privacy, the CNIL (French Data Protection Authority) could spearhead such cancellation proceedings.

And companies doing business in France should watch out—CNIL sanctions are no joke!

Are there specific events making you feel it has been neutered?

- Fentanyl is ~100 times more potent than morphine

- 2mg can kill

- Medical use is around 100ug/hour for patches

I had the same issue as you a few days ago. By separating the problem in smaller parts and addressing each parts one by one it got easier.

In your specific case I would try to fully complete the business logic one side. Reset the context. Then provide the logic to a new context and ask for an interface. Difficulty will arise when discovering that the logic is wrong or not suited to the UI, but i would keep using the same process to edit the code. Maybe two different contexts, one for logic, one for UI?

How did you do?

The regular backend stack on docker is smooth. Vs code / jetbrain is smooth. Nothing to complain about.

I could use more ports however. That is easily fixed with an USB hub.

Of course then there is the price of the subscription... But I'm talking about the model, not the cost.

Why not write simple SQL queries and use another language to do the transformations?

Are SQL engines really more efficient at filtering/matching/aggregating data when doing complex queries? Doesn't working without reusable blocks / tests / logs make development harder?

Syntax is one thing, but actual performance (and safety/maintenance) is another deal?

Do you enjoy going through your mail account and unsubscribe all useless newsletters? Use email aliases? are wary of all ads and brand claims you see on social media and prefer doing our own research? hate clicking on SEA results: usually the landing page is a weird, summed up version of the full website (we want the full thing)

But i guess that's a particularity of hackers and tinkerers :)

For many different reasons make people click and buy: discounted prices, FOMO, clever retargeting, funny ads, hidden ads (sponsored/influence/fake news),...