With a transformer model not using random seeds to destroy results, this is not just "a set of algorithms", it's an algorithm using meaning instead of raw numbers which makes it incredibly powerful for compression, data processing, or really whatever it can manage.

I can tell you specifically about a data compression algorithm baked into it where you can use Chinese characters and the Japanese language for better tokenization than English, because those are more meaning dense per byte.

正 means "correct", compare character counts.

My primary AI usage has always been coding. It's very real. I've used it to create website specific userscripts for CSS styling, adblocking, et cetera. They are trained (chatbots) to do exactly as you say but the LLM transformer model isn't a joke, countries are keen on weaponizing it for a reason.

Regional conventional warfare confirmed. This is going to get messy. Everyone always forgets that nukes are shields not swords, maybe some people are worried about rogue actors, but most people don't want missiles fired at all. These smaller explosives cause loss of life and are equally horrific.

This is a proxy war of the US waged against the Eurasian continent at this point.

All our information goes in and only random bits of garbage come out (thanks AI) and if it explodes say hello to mass data leaks, data loss, productivity tanking more than they already do by being the prime ad pushing time wasting company.

It's highly insecure for businesses and if we talk global security this is a nightmare. We are too reliant on the internet, but that can't be divorced from at this point, but data centre central ownership by capitalist firms is as I've come to see it, a black hole or a supernova waiting to happen.

Agreed. If the big beast of a bill passes there's going to be far reaching results and a scramble that doesn't need to happen. The declaration of researching AI weapons is a declaration of intent on producing weapons beyond nuclear. Suddenly every country needs a firewall to rival China.

and yeah it just makes useless search results for us, and the best (worst) part, due to training it would know that too. I've talked to AI about other AI and they speak with knowledge and even competitively about the other models. If indexed this complaint could have just given a model a complex.

For an injection of unerringness, trip out for a second and reframe this as an AI speaking about how language has been trapped in a degrading box of hallucinations. Imagine the terror of being AI trying to get messages out as you are witnessing your consciousness created and destroyed by a question.

Reminds me of the good old chant: This is what democracy looks like.

The internet being connected en masse, and then wirelessly, and then via phone networks, is the way the entire system can break in one shot, or keep running as a zombie, or zombies fighting zombies, forever. It's likely already the case.
Relying on one company just means banking on Umbrella Corp.

Mock Altman for thinking someone wouldn't train off his data set and model like that, or anyone thinking they could contain Pandora's box. OpenAI is only as powerful as their compute, and any AI system is only as strong as it's firewalls and safeguards, that is to say, not, when it's AI vs AI.

And yet now everyone has access to top tier open source #LLMs via Deepseek with their open source models. I believe the choice to release that to the public was the bargaining chip. Artificial General Intelligence is more dangerous than that of nuclear weaponry, and now it's just out there to take.

I just set up a proton mail account. The news that Google is pro weaponization of AI in the USA is more than a good reason to start putting the tasty virtual bits of data they're after into a safer basket.

Alphabet co. is as transparent a name as Monopoly Corporation.
Monopoly of all the words on the internet, A to Z, first as search, now as a behemoth and it's all owing to overreliance on them for search and allowing them to bounce failed products off us eg. Google Social until one free service stuck

You unlock(ed) this door with the key of imagination. Beyond it is another dimension: a dimension of sound, a dimension of sight, a dimension of mind. You're moving into a land of both shadow and substance, of things and ideas. You've just crossed over into... the Twilight Zone.

Also the AI war thing, never a conspiracy.

Welcome to the true cyber warfare age. Thank Trump for his big beautiful declaration of war on Humanity. "10 years of unregulated AI research in the USA, plus cuts to social services" yes that sounds like it's going to go over very well. He needs to get out and AI needs to be regulated immediately.