I wrote down some thoughts on foundation models as a public good and risks of bias https://andercot.substack.com/p/foundation-models-as-a-public-good what are on people's minds these days when it comes to the philosophy of large language models?