Yeah 55dB at 300m, maybe...

Yes, see above. That's exactly what the "temperature" variable sets.

No, it's technically wrong. Take an LLM, set temperature to 0, input a prompt, and it will produce exactly the same output every time. Hallucinations/confabulations are a separate question entirely, they speak to model utility/accuracy, not to whether it's deterministic or not.

Can we please stop saying that LLMs are 'nondeterministic'... It's misinformed. And it matters (because some people associate nondeterministism with some sort of magic, and others associate the word with inaccuracy -- neither of which are true). These models are clearly, architecturally, mathematically deterministic at a fundamental level, although this determinism is hidden from users. Let me explain. If you take any recent LL model (GPT, Llama, Gemini, etc...) and run it with 'temperature' set to zero, then, for a given input, it will produce exactly the same token, every single time it runs. That's not surprising, the weights are baked into the nodes of the model and calaculate the same output every time (of course excluding very rare things like uncorrected cosmic-ray bit-flips). Now, this has the side effect of introducing some undesirable behaviour (rigid responses, and looped outputs in certain contexts), so the 'temperature' parameter is used to mix things up a bit (this parameter is usually hidden ftom users in cloud models). When tokens are returrned on an inference run, there is an array of tokens provided with probabilities of being the 'right token'. The temperature value just uses a pseudorandom algorithm to return less likely tokens a proportion of the time (give the user the 2nd or 3rd most likely token some of the time instead of the most likely). To be clear, this has no effect whatsoever on the deterministic nature of the model, it's just a fudge-factor applied on the results after the token array is returned. The LLM models don't do something unpredictable or nondeterministic, they are just like any other computer algorithm run on a deterministic Turing machine. The debate around what sort of informational structure is represented in these models is interesting, but please don't lean on nondeterminism to explain your biases (positive ot negative about LLMs)... it's just incorrect.

Uh, the Trump administration is staffed by a bunch of people with openly anti-democratic views and they have strong opinions on how other countries should run their affairs as evidenced by J.D. Vance's speech at the Munich Security Conference. From a non-US, non-Chinese point of view, while China is anti-democratic, it is still better than the US in that China at least leaves you alone as long as you don't step on it's tail while the US will without any provocation try to force you to run your country the way they think you should, or outright annex your country like your orange king has threatened Canada and Greenland. For most of the world China is bad but the USA is actually becoming worse and that took some doing on the part of the US.

Salesforce Says: AI Customer Service Saves $100 Million Annually.
Salesforce Customers Say: Salesforce AI Customer Service is utterly, utterly useless.

That depends entirely upon whether US companies stand to benefit from EU anti-trust laws or not in any given situations. This is a far easier and more agile process in the US, you just keep buying Trump's meme-coin until he does what you want.

They do have a point about new players having trouble competing against abusive 'entrenched giants' but in the particular case of OpenAI I've got to say this is the pot calling the kettle black because OpenAI wants to become an abusive entrenched giant and if the way they have run roughshod over IP and copyrights to get where they are today isn't enough to convince the public of this then the public needs to be beaten over the head with a clue stick.

Charging stations are built and operated by profit making enterprises. One would think market forces would fix that problem. I'd be far more worried about political ideologues banning BEVs and charging stations because the are 'woke' technology, which is of course just an execute to artificially prop up oil companies and the dying ICE car industry instead of letting that fail which the market has already decided is doomed to failure.

The real question is what proportion of a day are nozzles actually occupied. I expect that number is extremely low. Chargers may take 30 minutes but they are just being used for a greater proportion of the day than gas nozzles. Furthermore, if there is twice the number of them than gas nozzles, tons of people also just charge at home overnight rather than using public charging stations and battery charge times are constantly decreasing I don't think a dearth of charging stations is going to be a problem unless Orange Palpatine declares electric transportation 'woke' passes a federal law banning charging stations and/or electric cars and then sends out police and the military to enforce it.

As long as they are license manufacturing chassis, plan to replace the software with domestic products and not buying knock down kits from China that should work out well for everybody ... except Americans, they'll be stuck with gigantic pickup trucks that start at 80.000 USD and underwhelming and overpriced ICE powered passenger cars due to protectionist import restrictions.

Unfortunately for every one of you there are at least ten others that behave the exact opposite way.

Maybe we should just rename "Germanium" to "Putinium"??

Nope, It's woke communism all the way down.

Unfortunately this is not an option for the US because 'recovering' Germanium is just another way of saying 'recycling' and that is both 'woke' and equivalent to practicing communism. Plus, Trump doesn't like Germany much so there's that too.