601 karma · joined December 16, 2019
That's just incompatible with modern life right?
I would imagine if you simply encoded the game in textual format and asked an LLM to come up with a series of moves, it would beat humans.
The problem here is more around perception than anything.
Very suspicious.
More like "Would you like to know more about XYZ, or circumstances that led to situation XYZ?"
They're already doing it, but wonder how far they'll take it.
Those are probably replies crafted by non-English speaking scammers from India / Russia / China.
There's probably a whole sea of undetectable replies from people who know how to prompt the models properly.
It costs hundreds of millions of dollars to train a frontier model. It's not just "scraping the web."
Distillation allows labs to replicate these results at 1/100th of the cost. This creates a prisoner's dilenmma which incentivizes labs to withhold their models from the public.
and a product manager