https://en.wikipedia.org/wiki/Simon_Oxley#Twitter_Bird_Logo_...

Reading this, I felt like I was too harsh on OpenAI for not monitoring their network, as they at least attempted to sandbox a workload.

I think in this case forcibly ejecting the injured ant could lead to more injuries of otherwise healthy ants.

[1]: https://www.sciencedirect.com/science/article/pii/S096098220...

Source: https://ec.europa.eu/eurostat/web/products-eurostat-news/w/d...

Comments URL: https://news.ycombinator.com/item?id=48672174

Points: 2

# Comments: 1

This seems like a complicated way to say "suckers". Of course they don't usually self-identify as such and think they act rationally.

In 2008 I was in high school and wrote a TI-BASIC tutorial in German [0] on my blog that became by far the most popular thing I wrote - maybe on par with my post about how to fix a quest bug in Skyrim by teleporting Delphine.

I was a bit mad back then that people for some reason appreciated those posts more than many very deep teenager ramblings about politics/philosophy :D

[0]: https://archive.haukeluebbers.de/2008/12/ti-basic-tutorial-1...

I have to say that intuitively I wasn't at all surprised that duplicating a single layer didn't do much good, but I had never expected that you can identify and so clearly visualize these relatively short circuit blocks (and of course it's around the magic number 7! /jk). Super cool research and really well explained!

But just now I read that a (sharply reduced) budget had passed the house? [2] does anyone know what the current state is?

[0]: https://www.csb.gov/assets/1/6/csb_cj_2026.pdf

[1]: https://www.pbs.org/newshour/show/only-federal-agency-that-i...

[2]: https://www.safetyandhealthmagazine.com/articles/27090-house...

Article in German: https://netzpolitik.org/2025/eu-ueberwachungsplaene-unionsfr...

Calling it "sideloading" instead of "installing" software successfully cements the notion that it is somehow not a completely normal thing to do. That's problem solved for the Googles and Apples of the world.

See the history of "jaywalking".

In my company some people are using LLMs to generate some of their code, but more are using them to get a first code review, before requesting a review by their colleagues.

This helps getting the easy/nitpicky stuff out of the way and thereby often saves us one feedback+fix cycle.

Examples would be "you changed this unit test, but didn't update the unit test name", "you changed this function but not the doc string", or "if you reorder these if statements you can avoid deep nesting". Nothing groundbreaking, but nice things.

We still review like we did before, but can often focus a little more on the "what" instead of the "how".

In this application, the LLM is kind of like a linter with fuzzy rules. We didn't stop reviewing code just because many languages come with standard formatters nowadays, either.

While the whole code generation aspect of AI is all the rage right now (and to quote the article):

> Focus on tangible changes in areas that you care about that really do seem connected to AI

Car producers can and do resolve this, e.g. iirc Audis don't use recuperation for the first breaking of the day. That way you don't have to remember to use the no-recuperation/break cleaning mode or break unnecessarily hard every now and then.

They exist for the reason to liquidate them in case of national emergencies/severe economic crises. It's easier to liquidate these reserves when they are stored in trading hubs. That could be New York and London, or maybe even Shanghai, if China wasn't a systemic rival.

Storing all of them at "one's place" is a larger risk than splitting it up and storing them in several places, each with a different risk profile.