codethief

You mean from ~8 months ago? https://www.anthropic.com/news/claude-opus-4-6

I would love to hear more about that, too!

> You set a goal. Agent will do goal.

So if I say the goal is to do X while not doing Y (e.g. breaking out of the sandbox), the agent will do anything to fulfill that goal to the letter?

Exactly. As someone much smarter than me once argued here on HN, the value of a software company is less in the software but much more in the mental map the team has built of the problem and of how the software solves it. Throw the team / mental map away and you'll have a very hard time maintaining the software.

Some OpenClaw/Hermes-like agents actually work like that. (Can't find the link right now.)

What a fantastic article! Thank you for sharing it!

> QEMU uses macOS hardware virtualization for the ARM64 desktop.

Huh, since when does QEMU support hardware virtualization on Apple Silicon? To the best of my knowledge, among all the VM managers available on macOS (VMware, Parallels, VirtualBox, …) only UTM currently supports hardware virtualization and that's through the Apple Virtualization backend, not through QEMU.

One could also phrase this principle in terms of entropy (in the information-theoretic sense): If I read some code and come across filter() and map(), I already have a pretty good idea of what the code is doing, without even looking at the filter criterion or the mapping function. Entropy is low. (There's only so many filter and map functions you could write for any given type or pair of types.)

Meanwhile, if I see reduce(), "anything" could happen. (Well, of course not anything but the set of possible reducers is surely much larger.) So entropy is high.

Avoid high-entropy constructs in your code. Try to keep entropy as low as possible. (For the same reason, code with a principled approach regarding side effects is a lot better than code where any function could mutate global state at any given time.)

> I had the chance to test it in Germany with Telefonica as it can be activated for free.

Interesting, where can I learn more about that?

EDIT: This looks promising: https://hilfe.o2online.de/o2-mobilfunknetz-einstellungen-zur...

Well I came here thinking this would be about https://en.wikipedia.org/wiki/Kernel-based_Virtual_Machine .

Agreed! Many people are saying AI isn't really intelligent yet because it can't come up with genuinely new things. Maybe finding a rigorous formulation of QFT / high-energy physics would be a great test for whether they are!

Maybe an opportunity for a hydroelectric power source?

> The “urine-no,” a hostile antiurination surface:

> Instead of minimising splash, it can instead be maximized by setting […], see Fig. 7a. Although not suitable as a commonly practical urinal, it showcases the design philosophy. Such a surface could be installed outdoors to deter public urination as the offender would fall prey to enhanced splash-back. This hostile surface may be dubbed as “urine-no,” as shown in Fig. 7b.

Love it!

I spent this past weekend giving UTM w/ Apple Virtualization a shot and can confirm that it really does work fantastically well! The only downside seems to be that it's not possible to pass USB devices through to the VM guest yet?

> but the latter is eroding very fast and they're roughly the same.

Citation needed. Some politicians keep on pushing for data harvesting and secret backdoors and what not but most attempts have been blocked and we're still nowhere near the level like, e.g., in the US.

Because they are nowhere near the same magnitude as in the US?

GrapheneOS has support for eSIM management these days but IIRC they need to invoke proprietary Google libraries for that to work.

> Notably, Hy4 preview also contributed to its own development process, participating for the first time in the automated optimization of training methods, data strategies, evaluation frameworks, and low-level operators. The model proposed approaches, ran experiments, and iterated based on the results, with the resulting code, logs, and feedback feeding into subsequent rounds of exploration. This established an early-stage recursive self-improvement loop.

This reminds me of one of the predictions from https://ai-2027.com/ . Only that there it's "OpenBrain" doing this, not the Chinese. And the authors of that paper were also slightly wrong about "Mid 2026: China Wakes Up": China woke up already a while ago. And:

> But China is falling behind on AI algorithms due to their weaker models. The Chinese intelligence agencies—among the best in the world—double down on their plans to steal OpenBrain’s weights.

No need to steal anything, they have already caught up.

And then there's this prediction for February 2027:

> Officials are most interested in its cyberwarfare capabilities: Agent-2 is “only” a little worse than the best human hackers

I think we're past that point now, too…

> The 10 dropped the physical SIM slot

Not sure about the 10, but at least the 10 Pro still has one.

> laboratory tests to determine the source of the infections were not expected to be available for several weeks

Note that this about DNA tests & analysis to determine the origin of the malaria/the mosquitos, not about simple tests for the disease.

There are great YouTube videos documenting this process btw. :-)

Malarone? Interesting, because that's supposed to be the prophylactic therapy with the mildest side effects. I have taken it many times and have never had any, and – up until now – I had also never heard of anyone having some.

> It's sad but unsurprising that no one was able to treat these German workers in time.

Literally every pharmacy over here sells treatment against malaria – many have it in stock and even if they don't, they'll be able to get it for you within a few hours. Every major city (certainly every city with an airport where you could catch airport malaria) has a hospital with a department specifically for infectious and tropical diseases, where they'll happily take care of you. GPs ask you if you've been to the equator lately if you go see them with high fever. I'm not sure how a country where malaria is not common at all would be able to do any better?

Also, as I mentioned in another comment, airport staff are instructed to watch out for malaria symptoms precisely because airport malaria is a thing.

Something definitely went wrong here but from afar it surely doesn't look like a healthcare failure to me. Maybe the deceased stayed home, thinking they had caught a cold and would be able to sit it out?

> It very glaringly points to some failure in the healthcare system.

Does it? "Have you been near the equator recently?" is a rather common question that GPs ask when you go see them with high fever. Doctors in Germany are absolutely aware of malaria, given how much we like to travel to all those beautiful, but unfortunately malaria-ridden countries.

The only thing, of course, is: That question wouldn't have helped here. However, this[0] source says airport staff are made aware of the possibility of contracting malaria and are told to watch out for symptoms.

[0]: https://www.tagesschau.de/inland/regional/hessen/malaria-am-...

Still works for me, too, and I haven't disabled any of Graphene's security features.

> There's a lot more broad cultural homogeneity between the countries in the Americas than there is within Europe.

That must be the most preposterous thing I've read in while, no offense! :-)

My personal experience has been the exact opposite: That cultural differences (and political differences no less!) in the Americas and also prejudices towards each other are huge, both between countries and often even between individual regions. (And that's even if you exclude the US for a second.) Also, countries in the Americas have had rather different histories (both pre and post Columbus) with only partial overlap. Meanwhile, European countries' history is practically impossible to disentangle, probably thanks in parts to how small the continent is in comparison to the Americas. On top of that, European culture has also had entire millenia to amalgamate and harmonize to the point where, yes, we may still speak different languages but it's very common for people to be able to trace back their ancestors' history across half the continent. It provides us all with the same cultural backdrop and the same roots as to where we come from, whereas in the case of the Americas I would ask: What is the common backdrop here? What are those common roots on a continent that combines both the incredibly rich history of hundreds of different indigenous peoples (all with their individual languages and customs) and the history of Europe, Africa and (in parts) Asia together? The partially shared experience of Spanish colonialism, and English, Spanish and Portuguese as shared languages only get you so far.

Finally, if your claim is that Europe is culturally much more heterogeneous (which again I don't think it is), I'd argue we must have exported a lot of that to the Americas, too. :-)

Emphasis on "the equivalent".

Further down he writes:

> I am of course using sanctioned cheating: I get all those tokens because I'm an individual, with the Claude Max discount. So it "only" costs me about $5k/month out of pocket, for a 50-agent cluster running on a 512GB M3 Ultra Mac Studio I bought off eBay for $25k.

> So it's not $120k/month of real money, but it's still crazy spend.

I've only skimmed the book and even though it was clearly written with the help of AI, there is some very good (and for the Python world novel) stuff in it. I particularly liked the chapter on effects.[0] I very much share the author's opinion that properly managing effects is the future:

> The history of programming is a history of scaling barriers. Each time, the pattern is the same. Something the programmer tracks by hand works fine in small programs. Systems grow until hand-tracking fails. The solution moves that tracking into the language or the toolchain, and a generation later, nobody can imagine doing it by hand. […] Effects are the barrier we are inside right now, which is why it is hard to see.

[0]: https://thinkinginpython.com/44_Effect_Management.html

Or street numbers? https://www.456bereastreet.com/ used to be a classic.

GrapheneOS heavily restricts JIT usage, too:

> - Android Runtime Just-In-Time (JIT) compilation/profiling is fully disabled and replaced with full ahead-of-time (AOT) compilation. The only JIT compilation in the base OS is the V8 JavaScript JIT which is disabled by default for the Vanadium browser with per-site exception support.

> - Dynamic code loading for both native code or Java/Kotlin classes is blocked for nearly the entire base OS. […]

> - Dynamic code loading for both native code or Java/Kotlin classes can be disabled for user installed apps via 3 exploit protection toggles: […]

https://grapheneos.org/features