em500

Also, employees do because it drives the promotion cycle.

You don't try to force them to match. You show a few very different implementations that all satisfy the underspecified the requirements. Not in an antagonistic way, but to help clarify and sharpen the requirements. If any implementation would do, then there's no problem.

The article claims that 6 individually named billionaires already left (presumably recently), and another is likely to leave if he loses his court challenge.

Tale as old as time (or at least as the 1970s): nobody ever got fired for buying ~~IBM~~ Cloudfare / AWS / Azure.

9 out of 10 corporate decisions are for blame avoidance / ass covering.

Note that almost all tests and marketing claims about M-DISC (most notably the 1000-year claim) where about the DVD versions. The main distinguishing characteristic of DVD M-DISC is that they do not use organic dyes like standard DVD+/-R, which can degrade relatively quickly. This is also why M-DISC DVD is not always compatible with any DVD reader/writer.

Verbatim / Millenniata loves to conflate claims about the original DVD M-DISC and the later BD-R M-DISC. But unlike DVD-R, standard BD-R already don't use organic dyes (with the exception of some low grade "LTH" discs). AFAIK there has never been good evidence that the BD-R M-DISCs have any special sauce compared to any other non-M-DISC BD-R. (Strengthened by the fact that there have never been compatibility issues or special equipment requirements to read/write M-DISC BD-R.)

It seems that Mistral is moving in that direction anyway:

"Our customers particularly value us for pioneering open models. Open weights give them what mission-critical work demands: the ability to see inside a model, adapt it, and retain the intelligence they build with it. This is why we are enthusiastic contributors to the Open Secure AI Alliance and NVIDIA Nemotron Coalition. We are now extending that openness beyond our own models. Mistral’s platform will support third-party open models, starting with Z.ai’s GLM-5.2."

[August 11, 2026] https://mistral.ai/news/regional-inference-open-models-new-c...

There are probably three main reasons: 1) unified memory (but you can also that with DGS Spark / Strix Halo), 2) access to your Apple account, so you can have a bot handle your iMessages, email, calendars 3) energy efficiency.

Apart from that, if it doesn't work out you still have a Mac Mini, which in itself is more desirable for many than a DGX Spark or Strix Halo if you have no AI use case.

I understood from informal chatter that researchers in the top Chinese labs are pretty open with sharing knowledge with each other. Additionally, it seems that anywhere between 30-50% of key researchers in the top US labs are ethnically Chinese. I wonder if this situation might give Chinese labs/researches some advantage just due to language and informal networks. Chinese researchers can understand all the English research, but research in Chinese is far less accessible to non-Chinese.

> From a cursory read of their white paper, my impression is that LISEP is trying to quantify the sense that making a livable income is harder and more disparate than it used to be.

But their own defined measure shows the opposite. The first graph on their website shows that their True Rate is currently at its lowest level over the past 30 years.

I wonder if someone already made a somewhat turnkey setup like this? Small models should ideally not waste a lot of parameters on trying to memorize too much knowledge and then hallucinate / interpolate a lot of the gaps.

I don't think this type of legislation will have any kind of real world effect. Apple App store labels all their buttons with "Get". Google Play Store just prints the price on the button for paid apps/games.

Everybody is in a bubble. Which is why it's worth looking into other people's bubbles occasionally.

https://www.pewresearch.org/global/2026/07/15/people-in-many...

Government issued versus corporate issued age verification is a false dichotomy. There are other options, such as refusing games that require them. (Yes, we do have a teen, and yes we did exactly that with Roblox.)

Why would that be a shame? "bc" is a mandatory POSIX command, while /bin/bash isn't (/bin/sh is the standard).

This article seems high on vibes, low on metrics.

> While a plain-vanilla philosophy degree remains as hard to monetize as ever, David Chalmers, a prominent philosopher of consciousness at N.Y.U., observes: “I think the demand for philosophers with A.I. training is, if anything, outstripping the supply right now. It’s an area I encourage students to go into. I think these issues with A.I. will be front and center for a good while.”

But wait, there's this:

> Beyond nonprofits like Eleos, most of the hiring has been concentrated at DeepMind and Anthropic, each of which employs at least a half-dozen philosophers.

So, between 6 and 12 each?

It's not "the company" that makes such decisions, but some senior managers or directors who do. They don't see a penny extra in their paycheck for savings money, but they are on the hook if anything goes wrong with less mainstream tech choices.

I think you're missing my point. China does not need to or aim to attract any researcher from anywhere. Just attracting top tier ethnic Chinese researchers and technical talent from the US is already a big unequivocal win for them. That talent pool is big: any US Math/Physics Olympiad team of the past 10-20 years is >2/3rds ethnically Chinese. And for them, Europe is not necessarily a more attractive destination than China.

It's not too different for ethnic Chinese researchers, of which there are a lot in American STEM departments.

Pertinent data: https://fred.stlouisfed.org/series/SPPOP65UPTOZSUSA

We've had the great small Qwen 3.6 early April that many could actually run on their laptop. Then similar from Google a few weeks later (Gemma4, better in prose, worse in code). Then the super cheap large Deepseek V4 a few weeks later. Then antirez DS4 build that made that actually runnable on MacBooks and Mac Studios. And now the "near-frontier / near-Opus" GLM 5.2.

For people who follow open LLMs, none of these were quiet and all were the most interesting open model release for a few days/weeks. In one or two months, it will be some other model again. Now I do appreciate the real rapid improvements in open models. But there's also a ton of hype and fast-fashion around all of this.

Just to be accurate, median annual salary is over $62k accross the US for kindergarten and elementary school teachers, with California nearing 100k.

[1] https://www.onetonline.org/link/localwages/25-2021.00?st=CA

Noteworthy that Z.ai, maker of the just released near-frontier GLM 5.2, has already been on the Entity List since Jan 2025[1]. Being on the Entity List does not mean all trade is forbidden. Broadly speaking it means American companies and individuals are not allowed sell them goods and services, but they are still allowed to buy from them and pay them.

AFAIK the Chinese AI companies barely depend on US goods and services, except for nVidia GPUs which were export restricted anyway, so it doesn't seem to be very consequential (see Z.ai). For the RAM maker CXMT it could be a lot more problematic though.

[1] https://en.wikipedia.org/wiki/Z.ai

Hourly wage for all private sector workers is up +32% since Dec 2019 ($37.54 vs $28.38)[1]. For non-management workers +35% ($32.31 in May 2026 vs $23.85 in Dec 2019)[2].

[1] https://fred.stlouisfed.org/series/CES0500000003

[2] https://fred.stlouisfed.org/series/AHETPI

> The biggest trap is the hallucinated citation. It will easily insert an absolutely authentic sounding quotation from another case that perfectly proves the point you are trying to make, then it'll make up an authentic name for it, e.g. United States v. Shenzhou Electronics Inc or whatever.

Naive question from an outsider: aren't there searchable databases of cases (with complete text) so that citations could be checked automatically, either by the same or an independent agent?

So this is specifically about Toyota's hybrid engines (which is very different from e.g. Honda's engine, not to mention plug-in hybrids or "mild hybrids"). The explanation of the mechanics starts around the 36 minute mark.

I always found most explanations of Toyota's Power Split Device too abstract, until I found this page where you can play with the sliders to see how the power is actually split between the ICE and the MG1/MG2 electric generators: https://eahart.com/prius/psd/

They learnt from ooenai that naming yourself open-xyz doesn't actually require opening anything.

> The cities' water usage is to keep people alive with basically mandatory things, like hygiene, and drinking

Almost half of city water usage is for residential landscape irrigation, mostly spraying lawns, which is not exactly mandatory or a basic necessity. Landscape irrigation uses about 3.5 million acre-feet / year, which is 1 to 2 order of magnitude higher than the estimated AI data center usage.

I sure hope connected health device makers include some generous tolerance for network outages.

> Why do you think an employer would waste resources like that?

The parent post specifically mentioned large organizations, where the "employer" is not some person who hires and pays employees from their own funds. Hiring and personel management is done by middle managers with their own interests and incentives, which can differ substantially from those of the owners or capital providers.

> The price of a loaf of bread in gold now is still similar to what it was in the Roman Empire 2000 years ago.

That seems pretty unlikely, when the price of a loaf of bread in gold now is less than half of what it was only 3 years ago.