It's easy to bash governments but, given that LLMs process basically all of human recorded knowledge and run on top of the collective efforts of 60 years of Moore's Law, we should look at them as a collective asset and need more good actors creating and improving them.
I can't picture any path forward that doesn't involve more customized versions with more agency in shaping them.
LLMs as we have them now are like mainframes compared to the PCs to follow but even more so as they have their own worldview and biases.
Good on Switzerland.
It's easy to bash governments but, given that LLMs process basically all of human recorded knowledge and run on top of the collective efforts of 60 years of Moore's Law, we should look at them as a collective asset and need more good actors creating and improving them.
I can't picture any path forward that doesn't involve more customized versions with more agency in shaping them.
LLMs as we have them now are like mainframes compared to the PCs to follow but even more so as they have their own worldview and biases.
The news is they finally published some models.
Get it here https://huggingface.co/collections/swiss-ai/apertus-llm-68b6...
You can try it here: https://chat.publicai.co
More discussion: https://news.ycombinator.com/item?id=45108401