4 ms·
The good news is if there are multiple frontier AI models from multiple countries with non overlapping sets of restricted answers, we can just use a couple of t
by epicureanideal 4mo ago
The good news is if there are multiple frontier AI models from multiple countries with non overlapping sets of restricted answers, we can just use a couple of them to get open answers.
- johnthedoe 4mo agoNot really non-overlapping though: both refuse to talk much about certain widely common activity between people (or even by yourself). That activity has shaped humanity quite a bit throughout its entire history. It's hard to imagine AI can understand humans fully if everything about it is excluded from the training data.
- paulddraper 4mo agoLimiting the output and excluding training data are not the same.