DrBazza

Physician heal thyself?

Tell an LLM to break into its own system as part of a feed back loop, but with no context of the initial creation? It should iteratively get to 'safe' shouldn't it (subject to the tokens you want to spend)?

For now yes. When LLMs can iterate and test at speeds humans cannot, it is going to continue to find holes.

An interesting side effect will be whether we accelerate moving towards memory safe languages (and development of more of those languages) in more sectors than just 'low-level'.

Not forgetting that this is only one mitigation of cyber-attacks and exploits and potentially removes a whole category of attacks developed by analyzing *source code*

All we need now is for Apple to reverse its choice to add the curved left-side sidebar (e.g. Safari), that when it pops out the window itself changes and the sidebar envelops/encloses the window's close/min/max buttons.

I click close at least once a day and close the window and not the sidebar that popped out. Windows, KDE, and Gnome don't do this.

'Find in page' is in.... 'Share' (on mobile). But Apple has the best UX, so who am I to argue?

I didn't once mention ethics, and I wonder why you think I did. Perhaps my last comment? That simply states, and nothing more, that there is a small number of powerful (if not rich) people in control of what might now be the largest empire on Earth today if you measure it by economic output and control (sorry USA), which is what the article talks about: a small group of people, in the 1800s British Empire. History repeating itself, but differently. Power is always concentrated.

Also the article starts with an implicit assumption of survivorship/observational bias, the title being 'Why were Victorian elites so effective?' - the assumption that they really were 'effective' or was the rest of the world simply worse and they were simply the 'least worst'? That line of enquiry was not even discussed.

No mention of survivorship bias, or observational bias - framing it as the most 'effective' empire of the time, rather than it was simply 'less bad' than the others. I'd like to have read about maybe the Belgians or French in comparison, and what they did that was 'less effective'.

It ends with this American-centric paragraph:

> Would modern societies work better if their local and national government were conducted by a tight-knit community of extremely rich and successful people? Maybe not: the downsides are obvious. But the problems would be interestingly different from those that they have in reality. Imagine a San Francisco whose Board of Supervisors featured Jeff Bezos, Larry Page, Sam Altman and Mark Zuckerberg, each feeling this responsibility to be the greatest glory of his career, each hoping that he might one day attain the greater glory that is the House of Representatives. It might not be paradise. Maybe it would be a disaster. But it would certainly be interesting.

Well, the Chinese Politburo is 21 people that run a country of (maybe) 1bn people. The progress of China and the lifestyles of a billion people over 40 years is pretty remarkable.

Because recruiters have turned it into their platform and if your industry requires recruiters, they're all now on linkedin and that's where you have to go. You can try getting a job via other routes, but connecting to recruiters, it's the lowest effort/least friction.

In a different reality, the UK government didn't allow 90% of ARM to be sold to Japan/SoftBank, and ARM expanded into chip fabrication in Europe.

The road to gigatonnes of CO2 is paved with the good intentions of the Greens in every country.

FWIW, a very quick look at other comparable sites (what seems to be the main css files):

sourcehut's 128kb raw, and 28kb over the wire.

codeberg is 420kb raw, and 66kb over the wire.

> but there is a difference between deterministic compilation and non-deterministic LLM code

But not the point of my comment. Computer programming has been a progression of physically wiring up valves, to soldering transistors, to punched cards, assembly, then higher level languages. Now we have natural language models.

The analogy being each that most people don't care about the assembly generated as the code works and it's really performant/efficient. Humans can still optimise assembly, but there's vanishingly small marginal gains for all but the most intensive/low-level tasks.

If LLMs produce things that work, and are indistinguishable from a careful human programmer (i.e. with some level of acceptable performance), people will simply stop looking at the high level code as the end result works, in the same way most people stopped looking at generated assembly after 8 bit computers (for example, as most games were written in raw assembly for... perforamance), as it was good enough.

> If instead I use an LLM to rewrite a feature of a codebase I can't be sure that it still functions like the original one.

Right now, with existing static analysis tooling, you can ask it to write a full suite of unit tests capturing existing behaviour without modifying the existing code with 100% code coverage, and start there. Plus fuzz tests as well. I actually have marginally more confidence in that than a human being doing it.

Very few people scrutinise assembly in 2026 as compiler generated code is 'good enough'. LLMs are beginning to do the same with higher level languages.

Without bashing anyone in particular, a certain OS-vendor's desktop apps, have been 'good enough' to ship, but with p*ss-poor performance in many cases for the last decade or so. We crossed the 'good enough' Rubicon a few years back in terms of what end users receive as a finished app.

Hopefully LLMs will eventually bridge that last gap of efficiency when generating higher-level code that not only works, but is efficient. Maybe there's a future where they generate the final binary without even invoking a compiler.