Daniel Paleka
@dpaleka
ai safety researcher | phd @CSatETH | https://danielpaleka.com
Recent LLM forecasters are getting better at predicting the future. But there's a challenge: How can we evaluate and compare AI forecasters without waiting years to see which predictions were right? (1/11)
my commentary on the IMO is that no sentient being, human or AI, should be subjected to solving P4. the outline of the solution is clear 5 minutes in, and then you spend like 90 minutes you will never get back trying not to fail at casework. otoh P1 and P5 are fantastic
you're laughing. you were born by chance during the final stage of The Great Work, the transmutation of all humanity’s work into a new form of life, the final boss of the anthropocene, the flower bloom of billions of years of biological life, and you’re laughing.
what is the most disappointing piece of software you use on a daily/weekly basis, purely on software quality terms
why is it that whenever i see survivorship bias on my timeline it already has the red-dotted plane in the replies?
*Goodfellas intro* As far back as I can remember, I always wanted to be an autist
and we shall see how great is the power of man in accumulating by his Selection successive slight variations
3.7 sonnet: *hands behind back* yes the tests do pass. why do you ask. what did you hear 4o: yes you are Jesus Christ's brother. now go. Nanjing awaits o3: Listen, sorry, I owe you a straight explanation. This was once revealed to me in a dream