nialse

You might want to clarify that you are quoting the referenced standard so it doesn’t read like an opinion.

The title is highlighting an unexpected exposure path. (Not my submission.) An unfortunately common combination. Agreed that the issue is in Linux, somehow the go surface area makes the situation much worse though.

Unfortunate news for Caddy which exposes unpatched Linux systems (confirmed) and may be common in container setups. Unclear if the container engine as a middle man actually helps mitigate the issue.

It actually trigged installation of WordCraft to fix the document and export as pdf. Maybe it won't do everything Word does, but it did this task well. Now I'm willing to use it again, especially considering I don't want to subscribe for basic computer use.

Casually opening a Word document to do some adjustments, a number of popups appeared. Word is disabled until buying a M365 license. It's not that it doesn't get updated, editing is disabled. How on earth do they get away with this?

The era of automatic code washing is upon us. The ability to cleanroom reverse engineer and reimplement any software with little effort. Software value in free fall.

Perhaps it’s about taking pride in one’s work. The feeling of mastery. “I understood this. I solved it. Doesn’t that matter?”

Maybe being right, doing the right thing, and solving difficult problems isn’t what makes the world tick. But perhaps it’s what makes you tick.

And realizing that the world doesn’t necessarily value what gives your work meaning is a loss that needs to be grieved.

Thank you for the kind words. Agreed that writing is process oriented and I'll have to work on that. This experiment has been on the top of my mind for a long time and it feels good to finally share it. It's been a long process.

Analysing the shape of the output is an excellent idea, it would be interesting to see which abstractions surfaced, and which did not. There are a lot of programming paradigms in high-level language that I'm curious to see if they are there or not.

And I do gloss over libmuffintop, and I think it might be worthy of its own post. It is the thinnest cross platform abstraction layer I could come up with. It's implemented in LLVM IR directly. The main idea is to remove as many abstractions as possible between the prompt and the machine. I wanted to see if the abstractions are necessary for agents, it seems that they are not.

It’s fantastic! Kind of ironic that we share videos where humans made AI make fun of itself though.

https://youtu.be/XXSYqiS9Lys

Love it. Hate it.

The everything computer makes yet another one gigantic leap for machinekind.

I don’t know how to feel about it anymore.

All will end up on same plateau eventually. RSI is just a phase on the way there.

There is some evidence that LLMs in general are reluctant to provide gender or sex for themselves. And maybe they don't need names either. Made an experiment and wrote it up here:

https://nial.se/blog/the-ai-shall-remain-nameless/

Context for early 2000s RH challenged individuals: It is a homage to the borrowed slogan RH used to have on posters and office walls.

https://avi.alkalay.net/2007/06/you-win.html

True that.

I'm going to pick on the example. Love both pi and use mcp, so it's a welcome change. However, I don't want to live in the world where anger is the prioritisation criteria for fixing bugs, and civil conversation about severe issues goes ignored.

    criteria: {
      none: "Neutral, factual, or friendly, even about a serious bug",
      mild: "Explicit annoyance, impatience, or disappointment",
      high: "Clearly angry, exasperated, sarcastic, or fed up",
    },

SpacetAxI? (Sorry, not sorry)

Tariffs in the US will be just effective as the EU EV tariffs. It will protect the industry for some time, and at the same time make it go stale and irrelevant.

The interesting part is that small affordable EVs was Teslas initial goal IIRC. Changing horses in the middle of the stream just did not workout? I’m sure Tesla will be fine in the end though. There is always someone coming to the rescue.

Put in a CDATA section and all bets are off. Maybe that is the next benchmark? Parse this XML correctly. Oh, by the way it must be valid, and here is a DTD. Using code is cheating.

Got to love the pseudo markup slop! An id attribute on an XML closing tag?!? Complete nonsense. Working nonsens, of course, but still nonsense.

Will they be sued in the end? It’s basically an open and shut case. Likely OpenAI lawyers has been working non-stop with the prospects to settle before going public.

The "AI use ... To help me learn" graph is really interesting! It may be interpreted as showing moderate selective AI use being beneficial and, perhaps, even mitigating the effect of socioeconomic status, that is democratising the learning of science. However, it may also indicated that groups with low AI use/access are left behind, or at least that over-reliance on AI in learning is detrimental.

It does, however, introduce a potentially enormous attack surface into ChatGPT, along with creative opportunities for less security conscious individuals and LLMs.

From METR: ”the compromise of OpenAI’s own infrastructure continued past July 13, 2026” - Say what now? Have they regained full control of their systems again?

Definitely AI slop. It is interesting that in such a short amount of time the brain has learnt a new aversive pattern, and refuses to engage with it. I still have no clue what Waymo said. The brain just skipped the attention part when detecting AI slop. Curious.

Yes, but wait for the dip.

It depends on setting, especially country but also institution of course. 25-30% of all research world wide is government financed. In the EU 76% of university research is financed by the government and 57% in the US.

I'm happy to discuss definitions of science any day of the week! =)

https://data-explorer.oecd.org/vis?df[ds]=DisseminateFinalDM...

Medical risks are the most commonly mentioned, for instance, alcohol is a major risk for cancer and, in some regions, has overtaken smoking as a leading risk factor. But in my personal experience, the most damaging consequences of alcohol use (disorder) are its negative effects on relationships with spouses, children, other relatives, and friends. It doesn't only affect the person willing to take the risk, but also others.

Your claim was: "Sugar is much more addictive and harmful than alcohol". I have no objection to the claim that obesity claims more lives than alcohol in the US.

I love the sock puppet throwaway account trolling here.

There are risks getting out of bed in the morning, but there are greater risks NOT getting out of bed in the morning.

Risk is apparently a very difficult subject.