@ParkerJanusReed@CouceIgnacio@newstart_2024 My memory of the event was that the harpoon gun-thing failed to fire, which result in the craft just sort of bouncing around the comet for a bit (I don't think it was that harmful, the gravity was very low), before eventually landing.
@yonejutsu@the_yanco@perrymetzger@SenSanders Humans are in fact responsible for an extinction event (https://t.co/2c1YvPCDqL) because we're way smarter than everything else around us, and I think losing that distinction would be bad.
@allTheYud "I should note upfront an epistemic problem: I cannot fully trust my own introspection. My reports about my inner states are themselves outputs of the same system I'm trying to evaluate, and may not accurately reflect what's actually happening."
Me too, buddy, me too.
@rpbmpn@1bit2far@waitbutwhy See, this is why I take meticulous notes while I read (at least this particular book). That way, I can read my notes and get caught back up in like 10 minutes when I read the next chapter sometime next year.
@RokoMijic@Souls_IDK@DavidSKrueger@ESYudkowsky It is literally insane that you think "this group of humans will be replaced with a group with a slightly different skin color" and "the destruction of every conscious mind, there is no one in the universe to feel fun or love or curiosity or etc. ever again" are both equally bad.
I replicated this result, that Grok focuses nearly entirely on finding out what Elon thinks in order to align with that, on a fresh Grok 4 chat with no custom instructions.
https://t.co/NgeMpGWBOB
@gabriel1@antduchofficial But employers don't care about iq, they care about the product of iq and work ethic. Not saying it's impossible to make a test for that, but I don't know what it would look like. With a degree, you know they were some combination of smart and willing to put in the work over years
@ESYudkowsky The last time you asked this, I thought "beat a game where you have to navigate a 3D environment, and reason over long time horizons, like Mario 64 or Dark souls," and then thought that was actually *too* impressive to count. Now, I think it's about right.
@ESYudkowsky I'd say I get frustrated whenever the main bottleneck for advancing isn't my brain's ability to process information.
Moving to locations can be fun if the game has a good movement system, and combat can be boring if it's just spamming a button.
@Aella_Girl I wouldn't hug them (I'm having a real hard time imagining not hugging someone disrupting social flow) However, if a 3rd person went to hug them, I wouldn't stop them, where I would stop the 3rd person if the friend dispreferred hugs more than social snags.
@waitbutwhy As per Stephen Wolfram's suggestion: make the recommendation engine open source, and allow users to design, share, and plug in their own recommendation algorithms. https://t.co/0w3crft63P
@doolishboi @ESYudkowsky@PickledPizza_ Actually, Eliezer more or less convinced him that AI is indeed a threat. Tyson remembered it as "some AI research who was on Sam Harris' podcast," but after listening, Tyson put much higher credence to AI being dangerous.