There isn’t a global law about age verification they countries could be exempt of. It’s individual countries doing it.

And on top of that the laws are different from what I’ve seen, in the UK for example you have to fullfil certain criteria to fall under that law. But frankly it seems to be a mess in my opinion.

This is what an age verification service says about it:

https://www.yoti.com/blog/understanding-age-verification-online-safety-act/

Sadly there is no answer for you available because many of the processes around this are hidden.

I can only chime in from my own amateur experiments and there are answer is a clear “depends”. Most adjustments are made either via additional training data. This simply means that you take more data and feed it indi an already trained LLM. The result is again an LLM black box with all its stochastic magic.

The other big way are system prompts. Those are simply instructions that already get interpreted as a part of te request and provide limitations.

These can get white fancy by now, in the sense of “when the following query asks you to count something run this python script with whatever you’re supposed to count as input, the result will be a json that you can take then and do XYZ with it.”

Or more simple: you tell the model to use other programs and how to use them.

For both approaches I don’t need to maintain list: For the first one I have no way of knowing what it’s doing in detail and I just need to keep the documents themselves.

For the second one it’s literally a human readable text.

The first part is a technical question and the second part a definition one.

For the how to: the most common approach is to simply blacklist their IPs on a provider basis. This leads to no provider that obeys your blacklists to allow their users traffic to that target. Usually all providers in a nation obey that nations law (I assume, I only know that for my own :D)

For the censorship: I don’t like that word because it’s implications fan be used against any and all laws. A shitload of content is made inaccessible because it breaks laws from active coordination of attacks to human trafficking. All of this can be described as censorship.

Forthe UK law it’s… I’m not British and to me it appears to be a vague tool to silence and control all types of content under the guise of protecting children. Not with the intention to protect or prevent something but with the intent to control. I would fully understand and emphasize with using the word censorship in this context.

Friend, there is only one kind of dumb question: those that are statements with a question marks at the end of are simply a vehicle to place a statement.

And even those are fine by me if the asking being realizes this within the discussion. (“Why are we allowing women to vote if they are that much dumber than men?” is a stupid question in my book - and even that can lead to “oh wow I always just assumed that to be true but (data/argument) made me realize that I was wrong!”. Rare but can happen…)

Worked for me, hope the copy paste does as well

Mysterious Antimatter Physics Discovered at the Large Hadron Collider

The LHCb experiment has observed a new difference between matter and antimatter in particles called baryons

By Clara Moskowitz edited by Lee Billings

Baryon particle illustration.

Thomas Parsons/Science Source

Matter and antimatter are like mirror opposites: they are the same in every respect except for their electric charge. Well, almost the same—very occasionally, matter and antimatter behave differently from each other, and when they do, physicists get very excited. Now scientists at the world’s largest particle collider have observed a new class of antimatter particles breaking down at a different rate than their matter counterparts. The discovery is a significant step in physicists’ quest to solve one of the biggest mysteries in the universe: why there is something rather than nothing.

The world around us is made of matter—the stars, planets, people and things that populate our cosmos are composed of atoms that contain only matter, and no antimatter. But it didn’t have to be this way. Our best theories suggest that when the universe was born it had equal amounts of matter and antimatter, and when the two made contact, they annihilated one another. For some reason, a small excess of matter survived and went on to create the physical world. Why? No one knows.

So physicists have been on the hunt for any sign of difference between matter and antimatter, known in the field as a violation of “charge conjugation–parity symmetry,” or CP violation, that could explain why some matter escaped destruction in the early universe.

Today physicists at the Large Hadron Collider (LHC)’s LHCb experiment published a paper in the journal Nature announcing that they’ve measured CP violation for the first time in baryons—the class of particles that includes the protons and neutrons inside atoms. Baryons are all built from triplets of even smaller particles called quarks. Previous experiments dating back to 1964 had seen CP violation in meson particles, which unlike baryons are made of a quark-antiquark pair. In the new experiment, scientists observed that baryons made of an up quark, a down quark and one of their more exotic cousins called a beauty quark decay more often than baryons made of the antimatter versions of those same three quarks.

Magnet for the LHCb (large hadron collider beauty) particle detector at CERN (the European particle physics laboratory) near Geneva, Switzerland.

CERN/Science Source

“This is a milestone in the search for CP violation,” says Xueting Yang of Peking University, a member of the LHCb team that analyzed the data behind the measurement. “Since baryons are the building blocks of the everyday things around us, the first observation of CP violation in baryons opens a new window for us to search for hints of new physics.”

The LHCb experiment is the only machine in the world that can summon sufficient energies to make baryons containing beauty quarks. It does this by accelerating protons to nearly the speed of light, then smashing them together in about 200 million collisions every second. As the protons dissolve, the energy of the crash springs new particles into being.

“It is an amazing measurement,” says theoretical physicist Edward Witten of the Institute for Advanced Study, who was not involved in the experiment. "Baryons containing b [beauty] quarks are relatively hard to produce, and CP violation is very delicate and hard to study.”

The 69-foot-long, 6,000-ton LHCb experiment can track all the particles created during the collisions and the many different ways they can break down into smaller particles. “The detector is like a gigantic four-dimensional camera that is able to record the passage of all the particles through it,” says LHCb spokesperson and study co-author Vincenzo Vagnoni of the Italian National Institute of Nuclear Physics (INFN). “With all this information, we can reconstruct precisely what happened in the initial collision and everything that came out and then decayed.”

The matter-antimatter difference scientists observed in this case is relatively small, and it fits within predictions of the Standard Model of particle physics—the reigning theory of the subatomic realm. This puny amount of CP violation, however, cannot account for the profound asymmetry between matter and antimatter we see throughout space.

“The measurement itself is a great achievement, but the result, to me, is not surprising,” says Jessica Turner, a theoretical physicist at Durham University in England, who was not involved in the research. “The observed CP violation seems to be in line with what has been measured before in the quark sector, and we know that is not enough to produce the observed baryon asymmetry.”

To understand how matter got the upper hand in the early universe, physicists must find new ways that matter and antimatter diverge, most likely via particles that have yet to be seen. “There should be a new class of particles that were present in the early universe, which exhibit a much larger amount of this behavior,” Vagnoni says. “We are trying to find little discrepancies between what we observe and what is predicted by the Standard Model. If we find a discrepancy, then we can pinpoint what is wrong.”

The researchers hope to discover more cracks in the Standard Model as the experiment keeps running. Eventually LHCb should collect about 30 times more data than was used for this analysis, which will allow physicists to search for CP violation in particle decays that are even rarer than the one observed here. So stay tuned for an answer to why anything exists at all.

It’s a question of effort. Sony has a shitload of public presence. For social engineering I can learn many mid level manager names from LinkedIn for example and their infrastructure is necessarily public facing to allow people to work there.

And that’s not talking about their public web presence and services.

And now we’ll switch to … You! If I’d try to target you I would have to first find anything from you to actually target.

Once I have your phone number, public IP or anything that gives me a lead I have to find my way in. And that way in will be because you’ve made a mistake, are lax with your passwords or use an out of date service.

But that’s like 2/3 of the work I had for Sony as well. And now I see that you’re a student with a net fortune of 50$ and a car from 1989.

To out it another way: for companies I aim with s rifle as they are a worthy prey. For individual people I use a shotgun and hope something hits something.

German here: just creating and selling something is one thing that jumps to my mind.

The concept of “I have an idea and a bit of money so I’ll just found a company” is … Tiresome. Possible, yes, but the legal hurdles both good and bad are ridiculous. You need way more time than in the US just for the formal overhead and even then you are way more in it with your own private existence.

As founder “beschränkte Haftung” is not as limited as it sounds at first if you’re not firm in legalese for example.

Not the OP so their point of view might differ.

I’ve only seen LLM and ADHD connected via writing / homework tasks. Perhaps that’s the same link OP thought of.

And the txt in the image is anyway just an aggressive individual opinion. I don’t get the logic of “I don’t want or use this so no one is allowed to see it as beneficial”.

You got a lot of relevant answers so I want to point out something else:

You’re hosting your own services. By yourself. Fuck everyone with a broom who tries to gatekeep that. And I don’t mean wooden side first.

Seriously, your question is on point here from my perspective and as long as it has a connection to running services by your own I personally would love more diversity in hosting solutions.

Personally, I’d love to see people share more about their provider agnostic opentofu deployment or someone who went all in on AWS lambdas for weird stuff.

It sorts by what seems to me historically by relevance, i.e. which day is asked more often because it seems a more frequent timeframe for everyday use in a medieval society compared to the month (with the seasons as something in between those two).

And I agree that since the digital age yyyy-mm–dd has significant advantages!

It really becomes time that it’s been made clear about that! It’s unbelievable that it took so long for someone to call out that it’s the time the It is referring to!

(Now I wait for some nerd to clear up each reference point because I only wrote that because I had the first two parte in my head when I reset your reply!)

Oh that’s an approach I haven’t even considered, I es always very … Black and white :D

Thanks for the inspiration, should I find a bucket of time in the future I’ll give it a shot!

Not the person you asked but for me:

It’s because they’re a documentation mess caught in a release/stable discussion that I don’t understand.

Perhaps that changed in the last year, haven’t looked into them again since then - but unless there’s a trigger (new server setup or official release for example) I simply don’t have the time and energy to invest.

I really like the question, thank you! The answer is a clear “yes, but”:

Your assumption is absolutely correct, light surfaces reflect more light back and heat up less because of this.

Noe if your display heats up less depends on the amount of energy it uses to generate that white:

For an e ink display it would be basically the same as a bright vs dark paper of the same color. But OLED for example uses constant energy to generate the white image: So it’s depending on how bright the sun shines vs how much heat gets generated by the display itself.

Still only looking at the sun’s energy it would be smaller. If the overall temperature would be lower depends on exactly how bright the sun would be vs how efficient the specific display is!

Raising it lowering the container works only if it is not protected from environmental pressure - but you’re right: creating two different pressure environments that are somehow connected is ready in a thought experiment but then you tell that to the engineer and they get their third heart attack of the week…

I’m writing only based on your text, not the video, please excuse any doubling of content.

It is easier explained if you build an imaginary machine instead of lifting / lowering that does the same thing. The single most important thing to understand is that the lower the pressure the less heat you need to add to boil something. There are funny graphs for each liquid (for example https://courses.lumenlearning.com/umes-cheminter/chapter/vapor-pressure-curves/ ).

The intro explanation

The water in your containers will behave based on their individual combination of pressure and temperature. I’d at any point the water vapor falls below its boiling point at the current pressure it starts to form a liquid. At this point you’ve made a fancy rain machine.

Note that water itself adds pressure to a system because of its volume even as a gas

A machine

Imagine you have a container at 100 mmHg which according to a random online calculator leads to a boiling temperature of 50 degrees C.

Now you heat this up and lead the water vapor into another chamber which has only s pressure of 10 mmHg. Water has a boiling temperature of only a bit over 10C there! So you keep it at 20C to be sure the water never gets liquid again.

But wait: now you’re adding water vapor into a low pressure container - you’re literally pressing a gas into it - so you increase the pressure in there.

The first container, the source of the gas, becomes irrelevant: As soon as the additional water increases the pressure to around 20mmHg it starts condensation again as now it’s boiling point moved above the 20 degrees.

The flaws

As you’ve asked for the downsides: it’s a very convoluted way of manipulating water to achieve the same result as simply heating it. You would need way more energy to lift the containers far enough or otherwise decrease the pressure than the energy needed to boil it.

Other than energy and logistics I don’t see a downside. Liquids don’t behave differently in terms of boiling no matter the source: pressure, temperature or a combination.

In this reply you you talked about “some breads”, the OP Post only talks about bread - and that for sure had only ingredients in using at home.

Same for French fries: potato, salt, fat .

I’m with the poor downvoted fellow, I don’t understand where the risk comes from when it’s described this vague.

Are home made burgers better? Is it the freezing process and I should lower my meal prep? Is it additives?

Hey,

I’d be very grateful if you could share your approach den if it’s only to compare (I went with a “be assertive and clear, skip all overhead” system prompt.

This is not only interesting for chatgpt but understanding how people solve these issues comes in handy when switching to local variants as well!

Thanks in advance

The impulse to start talking or interrupting people you can pin to ADHD no problem - but tunnel firing for a whole dialogue discussing a problem?

Yes there is but it’s because of German law in they regard. And as it has to be approved by a governance body in public there is no chance of obfuscating it.

Normally that is a good thing from my point of view as it prevent shady weapons deals - but this is not a normal situation :(