My impression is that the US is more vulnerable to climate change than Europe. While Europe has heat waves, the US has a hurricane problem that simply does not exist in Europe, which can result in horrendous societal damage.
“We used LLM technology, which is great at parroting content, to attempt to predict what the US president’s spokesperson would say at their next conference.
We used that as input for ~gambling~ purchasing a position on a prediction market, which has been popularized recently in part due to its ability to circumvent gambling regulations.
However, even the LLM couldn’t parrot the words of the spokesperson. The implication is that the spokesperson speaks so outrageously that even an uncensored LLM couldn’t parrot their words.”
I try to stay humble when predicting the future. But there is just no way there will be a literal military invasion. Trump would never risk a bunch of american dying on the ground, it would be terrible optics.
Maybe they just want access to the specific companies? Like, you get hired by the company. You hand over username/password/vpn-info to them. Then they have a way inside the company and can try to steal information, install backdoors, whatever, with very low risk of getting caught.
I added the custom instruction "Please go straight to the point, be less chatty". Now it begins every answer with: "Straight to the point, no fluff:" or something similar. It seems to be perfectly unable to simply write out the answer without some form of small talk first.
Maybe until the model outputs some affirming preamble, it’s still somewhat probable that it might disagree with the user’s request? So the agreement fluff is kind of like it making the decision to heed the request. Especially if we the consider tokens as the medium by which the model “thinks”. Not to anthropomorphize the damn things too much.
Also I wonder if it could be a side effect of all the supposed alignment efforts that go into training. If you train in a bunch of negative reinforcement samples where the model says something like “sorry I can’t do that” maybe it pushes the model to say things like “sure I’ll do that” in positive cases too?
I had a similar instruction and in voice mode I had it trying to make a story for a game that my daughter and I were playing where it would occasionally say “3,2,1 go!” or perhaps throw us off and say “3,2,1, snow!” or other rhymes.
Long story short it took me a while to figure out why I had to keep telling it to keep going and the story was so straightforward.
I guess because it is full of guesswork and devoid of real factual research (at least for the main headline question). But it turns out that bloggers looking for content and lacking any skill are also capable of writing plausible-sounding slop.
