Natanael

This gets at my own personal perspective of using LLMs to respond - it’s not just about not putting effort into understanding and responding yourself, rather it is about making yourself a proxy to a tool I could use myself, and doing so *without even having a better understanding of how to use the tool to answer my question*, and still thinking you’re somehow made a positive contribution, that is the most disrespectful.

If you genuinely thought the LLM could help me then you should be explaining your process to me for how to use it and validate responses, or else at least you should ask me for more info and explain how you think it’s responses could help if you really do think you’re better at operating it.

Imagine doing the same in a workshop, and taking a powertool to an object before you even bothered figuring out what the other person wanted. Or trying to be helpful by asking questions on your behalf to other departments, but messing up the context and thus repeatedly producing useless answers that you have to put time into refuting.

https://www.lawfaremedia.org/article/what-the-supreme-court-got-wrong-in-the-trump-section-3-case

SCOTUS didn’t even rule on if he’s disqualified.

https://reason.com/volokh/2025/05/20/judge-rules-removal-of-u-s-institute-of-peace-usip-directors-was-illegal/

It’s not the same type of case, but it’s relevant to show total legal nullification and total reversal is legally supported. It can be extended to the entire second term, erasing every change he made.

[T]he purported removal of Ambassador George Moose as acting president of the Institute by a resolution adopted by less than a majority of the duly appointed Board of Directors of USIP was invalid, and therefore null, void, and without legal effect ….

Enforce 14A3 and declare the second term invalid and null and void. That will nullify all the pardons too

Yes.

Meanwhile intelligence agencies says they want to stop encryption to help them investigate crime…

Lobbyists selected by rich people tell politicians who to appoint to lead agencies. These people give directives and can get agents fired for disobeying.

Stopping it means removing “single points of failures” like not having such a powerful hierarchy, in addition to accountability measures (independent inspections, etc)

Presidential pardons for state crimes don’t exist. Claiming he’s done it does nothing. Signing fancy formal papers saying so is null and void.

Every time the policies and suggestions and everything else was presented without names, Kamala’s campaign won. When people didn’t know who’s proposal they was listening to, Kamala won.

So please explain what you mean about better candidate, because you have said nothing about how she was worse.

From here in Sweden, it looks like the things needing replacement is the fucking electorate and the billionaires owning the media companies

What part of the campaign was worse

Please specify

Did you not pay attention to how Kamala was called a liar by media for saying Trump would send the military against cities?

… And then he did?

Did you not pay attention to how he claimed credit for Covid funds that Democrats established? Blaming Democrats for Trump judges’ decisions? Setting up Afghanistan so the exit would happen at the start of Biden’s turn to hurt his approval when it predictably turns to chaos? How Biden’s age was somehow a problem but Trump’s never was?

Media was hysterical about how democrats would make everything worse somehow but never explaining it, then repeating Trump’s claims that he’s deliver the best economy ever and all kinds of bullshit like it.

And with all that Trump still didn’t even get 50%!

The question is if it wins then more support than they lose.

It’s actually kinda easy. Neural networks are just weirder than usual logic gate circuits. You can program them just the same and insert explicit controlled logic and deterministic behavior. To somebody who don’t know the details of LLM training, they wouldn’t be able to tell much of a difference. It will be packaged as a bundle of node weights and work with the same interfaces and all.

The reason that doesn’t work well if you try to insert strict logic into a traditional LLM despite the node properties being well known is because of how intricately interwoven and mutually dependent all the different parts of the network is (that’s why it’s a LARGE language model). You can’t just arbitrarily edit anything or insert more nodes or replace logic, you don’t know what you might break. It’s easier to place inserted logic outside of the LLM network and train the model to interact with it (“tool use”).

If you asked people about their policies and literally absolutely anything they wanted to do, and hid the names, Trump always lost because he is MORE unpopular

Media helped Trump steal credit for things others did, instill fear of things his opponents never would do, etc. Nobody lost to Trump by being a worse candidate, but because media refused to be honest and people believed the lies

ICE already says their facial recognition app is more accurate than your actual citizenship papers.

Trump lost the popular vote in 2016 and 2020 and still was below 50% in 2024 AND his policies never got support from a majority ever when presented without a name.

There has never been a majority who has liked anything he had to offer.

This was a propaganda war from racist wealthy assholes who bought out a ton of the companies controlling the media landscape. They exploited bullshit election rules and lied and bribed and extorted to suppress resistance, all just so they could be assholes in public without being called in it.

People who opposed the monster won’t feel offended by that. Normies won’t be here to see it.

And some of these hardline ultrapurists has to be shamed into doing the right thing

Use it as a dumb worklog sometimes. Tell it what you were already gonna do and ask if it would do the same. It’s almost always gonna agree. Then just ignore it. If somebody AI obsessed pulls out the full logs they’re gonna see you’re doing what the AI said was good. (basically Inception, lol)

Here in Sweden, I’m pretty sure pills are only distributed individually like that in controlled settings (elder care facilities, etc). Otherwise you just get the whole box.

Depends on the provider. Many only allow a single use of the provisioning code.

Some providers does however let you create a new one whenever (meant to be used when you replace devices)

Gmail started at 1 GB, which was massive compared to anything from 1-20 MB at most hosts, and they made a point of storage per account growing over time until they jumped to 10 GB

Some Sony phones have that type of sensor