Qwen 3.5 can be run via ollama

Qwen 3.5 is one of the best of the open-weight (self-host able) models right now. It’s not as good as some of the extra massive proprietary models like the bigger Claude models.

Sure, but to get the communication started you would start with facts you’d agree on, like the positions of stars or basic chemistry.

The model we currently have for the universe goes well beyond anything we could learn with our natural senses and the way we intuitively think about the world because of those senses.

It’s true that we keep refining our models an it’s ver possible that an alien would have slightly different models, but at the end of the day, we are trying to describe the same universe and those models are going to overlap a lot because of that.

First of all, there has been a lot of research into what the minimal set of assumptions you need is to reproduce what we consider “basic math” and also what happens if you tweak those assumptions.

Second of all, the main goal for science and the type of math we use for science is to effectively model the world we live in.

Any aliens that live in the same universe are subject to the same physics, and any civilization advanced enough to detect our messages will know some basic universal facts about the world, and those facts are what we hope to use as the basis for starting communication.

data security in that case had nothing to do with the llm

That’s kinda my point.

“I don’t trust companies to hold their promises” is a very different argument from:

LLMs are inherently bad at data security and there is no way these companies can, in good faith, promise HIPPA compliance

It is certainly possible to implement a secure LLM service.

High physical fitness is a requirement and that helps with the jaw line

Phone, wallet, keys, and headphones

This is about extracting data that was used as training data. Just don’t do that with sensitive data.

LLMs are inherently bad at data security and there is no way these companies can, in good faith, promise HIPPA compliance

This is simply false. AI sucks but it doesn’t help to lie about it.

EDIT:

Go run a local model on your own computer, and delete the context when you are done. Boom you just used an LLM in a way that maintains your data security.