pphysch

Factorio is in that top-tier. I just tried Core Keeper and had the same feeling, gushing with QoL and thoughtful design. Refreshing to play sandbox-y games where mods don't feel mandatory.

Aside from being obviously written by a LLM, this scenario reminds me of the "LLM, say you're alive"; "I'm alive!"; "Oh my god..." meme.

You're allowing the attacker not only direct access to modify the source material, but giving them multiple informed attempts/turns at optimizing the output in their favor. This is like a worst-case insider attack; what systems are supposed to be resilient to an undetected attacker with 'root' access?

This would be slightly more interesting if the attacker could consistently one-shot the task, but it's taking half a dozen attempts...

The extraneous small-caps labels are one of the main idiosyncrasies of AI generated markup. I wonder how much of this is a "scaffolding" technique to help the model build stable designs. But was it reinforced in RLHF or an emergent behavior of the models?

According to the other side.

What was "military" about those motorboats? You're suggesting the same military has vision into dozens of little boat's barrel contents but can't see hundreds of children in a school yard?

How can anyone take this argument seriously when the DoW was simultaneously committing dozens of extrajudicial executions of small boat users in the Gulf?

It's called a nickname, and it's in reference to her infamous article defending the murder of actual journalists (which she is not).

Can you answer my question?

"Propagandist" is not a palatable job title, so propagandists often masquerade under different words. For example, the US government's official foreign propaganda agency is called "US Agency for Global Media".

If Anne "it's okay to kill journalists in wartime" Applebaum is not a prominent Western propagandist, then who is?

Is it easy? What if an organization avoids liability through shell companies and financial engineering?

"The company managing your warranty was acquired by Foobar Inc and their sole building burned down destroying all records. We apologize for the inconvenience."

"brazen and identified" is doing a lot of work here. IIRC the cable thing was an anchor accident, and everyone is doing psyop campaigns (Western psyops being far more advanced and sophisticated than their enemies). Nord Stream was definitely not an accident and we know who did that.

Implying that Russia somehow invented or is the undisputed leader in "psyops" is a popular hallmark of a low-information civilian, a narrative promoted by arch-propagandists like Peter Pomerantsev and Anne Applebaum.

As many other comments have said, cui bono?

Never attribute to delusion* what can be adequately attributed to a pre-IPO hype campaign.

* or anything, really.

A vending machine in Anthropic's office, which stands to gain from narratives like "I used Claude to generate profit".

This is a surprising development. I wonder how much of it was opportunistic, since Talos 1.14 added sandbox isolation based on Linux namespaces (?). Perhaps the reasoning was "let's just go one step further and support full containers/VMs instead of inventing our own sandbox mechanism".

That's a bizarre take. Which of Israel or Canada has assaulted a US battleship unprovoked?

The comment I am responding to is attempting to spin it as some kind of authentic concern rather than more IPO hype building.

This is the CEO of a company with an upcoming IPO publicly saying "my product is potent and valuable".

We should not put any spin on it. There is nothing more to it.

That's very hypothetical, but one might also ponder what under scenario Israel might attempt to sink a US battleship and murder dozens of US servicemen.

It's hard to predict what a theocracy might do.

I suggest you look into the many confirmed impacts in Israel, including at heavily protected airbases and downtown Tel Aviv where the regime embeds its military intelligence centers.

There's a reason why Israel has stopped directly striking Iran.

> Maybe we should have brokered some kind of deal with Russia to prevent them from invading Ukraine.

I understand history and contemporary events is not your strong suit, but such a deal was very much on the table in December 2021 and the Biden Admin laughed it off. Then again in April 2022.

"Sure, Israel might be mass-murdering children while calling them 'Amalek', but at least they aren't scary Muslims!"

At what point does reality become more important than your favorite childhood stories?

Iran clearly has the capability to flatten every major hospital and university and desalination plant in Israel, like Israel has done in Gaza, but they don't.

It's also clear that Iran having nukes would be a strategic deterrent to Israel's ultraviolent territorial expansionism.

[flagged]

The USA is heavily supporting Ukraine with intelligence sharing, private financial support, StarLink, and """indirectly""" providing arms through European middlemen.

The latter was Trump's innovation so that he can pretend he is upholding some foreign policy campaign promise without materially changing anything.

The two major ongoing wars are due to US foreign policy and could be over tomorrow if US government fully stopped supporting their respective belligerents.

A good chunk of the human population has never known a "Pax Americana" and regions such as East Asia are clearly better off due to declining American hegemony.

Who is taking it seriously? The unipolarists clearly don't, and the multipolarists have been busy building alternatives for the past couple decades.

This is straight up misleading. When you press your K key on a touchscreen, your keyboard program may decide you meant to press the neighboring L key (by dynamically inflating the collision geometry on it) because it was statistically far more likely that you meant to press the L key next.

This likely doesn't happen exactly on an analog keyboard, but then many text-processing environments that do the same thing in post. My keyboard just edited 'yuor' to 'your' even though I successfully input the prior string.

Yes, and keyboards also save a lot of time over handwriting. Numerical methods and proof engines save even more time. LLMs are just another tool in the kit.

That's usually a good sign you are on shaky ground!

In the same way you rely on a keyboard or touchscreen to type this comment. It doesn't mean the tool is the brain behind the work.

Nvidia is among the best positioned if open models win.