Bluesky Wrapped

As a general rule, an LLM can't reflect reliably on its own process. Any explanations after the fact are generated in the same way as the original response. You press for an answer. It responds like a human would answer - not reflecting on the actual process. So you end up with endless apologies.

We share the sentiment in Denmark. If I see one more pundit smile or laugh when they mention plans to annex Greenland or Canada, I will lose it I know most US Americans find it wildly unrealistic, but the US has done this for decades. We just never thought we could be targets. That's on us

What's most interesting here is the fact that ChatGPT has picked up the (fairly recent) over-use of the "DEI hire" phrase. It shows how human language flows naturally into the LLMs as they absorb new content. Also, the importance of training on balanced inputs...