BeefWithTheChief

AI Superintelligence Research!

[BeefWithTheChief @HesNotABaby1]

@lukeburgis There's a bit of a selection bias here. If we are in a position to look back on this in a few years from now, it means they were wrong. If they are right...

[BeefWithTheChief @HesNotABaby1]

@pileofrocks3 @robbensinger Greg Casar read the METR report and has a plan to get a bipartisan ban for RSI, which I imagine just kicks the can compared to a full pause but is a lot easier to imagine happening than a whole pack of R's signing onto Ban AI Superintelligence Act in Jan https://t.co/D7lxycO3wR

[BeefWithTheChief @HesNotABaby1]

@tedlieu We need a pause or we don't make it out of 2027

[BeefWithTheChief @HesNotABaby1]

@tszzl Roon, I feel threatened by the computer.

[BeefWithTheChief @HesNotABaby1]

@DavidSKrueger Thinking about this, it would also help if there was like a list of Congresspeople who were on the fence who we could campaign for, the "easily be nudged" category

[BeefWithTheChief @HesNotABaby1]

@MTSlive @eli_lifland What a headline, you literally asked him "What if you're wrong, how would you best explain that?" rather than whether he thinks he's wrong

[BeefWithTheChief @HesNotABaby1]

@MTSlive @ramez Let's put it this way, I haven't seen an in depth assessment of the METR investigation that -hasn't- put superintelligence in the next year. Has Yudkowsky put something in all our wells?

[BeefWithTheChief @HesNotABaby1]

@MTSlive "in 10 years, cancer won't be a thing you won't have to worry about anymore" - SNL Dario

[BeefWithTheChief @HesNotABaby1]

@robbensinger @richardhutton Thank you, I don't know how you are so brave but you're good at talking people down from the ledge

[BeefWithTheChief @HesNotABaby1]

@DavidSKrueger I hope to heaven you are right! Bernie Sanders/Greg Casar's Ban AI Superintelligence Act seems like a good place to start, right?

[BeefWithTheChief @HesNotABaby1]

@robbensinger @richardhutton Rob, after the AIFP report I don't see a way forward. I would do anything for my loved ones but I'm no longer seeing it.

[BeefWithTheChief @HesNotABaby1]

@RepDelBene "Pausing AI" just means the pause of capabilities and growth towards existentially dangerous superintelligence. We can still work on alignment (so we can grow capabilities) and cancer cures etc.! If we don't pause, we don't survive.

[BeefWithTheChief @HesNotABaby1]

@JBPritzker What Pritzker is calling for is insufficient. We need an AI pause immediately, since we seem to be on track for an AI takeover in ~1 year, as per extrapolations on METR's report on Anthropic last week.

[BeefWithTheChief @HesNotABaby1]

@heartwr3nched @JBPritzker An independent report by METR has come out that shows that Anthropic has reached recursive self improvement. When the data is extrapolated on, it looks like we have ~1 year until the AI takeover. I'm not sure if Pritzker knows or cares but that's why X is in a frenzy

[BeefWithTheChief @HesNotABaby1]

It's high time for a Daisy ad!

[Do Not Accept @DONOTACCEPTai]

AI companies admit what they’re building might kill everyone. We don't have to agree to this.

Add your name at https://t.co/Pcg0clMbfV to say: we do not accept your terms.

[BeefWithTheChief @HesNotABaby1]

@So8res Daisy ad? I'm down

[BeefWithTheChief @HesNotABaby1]

@dewierwan_ The only good thing about their plan, and the part that I wish we could extract in isolation from the human extinction, is the part where they go to a different galaxy from us.

[BeefWithTheChief @HesNotABaby1]

If I hear one more time of a Democrat saying that AI is so dangerous for us all and then referring to "sensible/commonsense/responsible regulations," namely, their plan to pass liability laws. ("Oop! We're extinct! Time to sue!")

[Stephanie H. Murray @stephmurrayyyy]

At this point, it feels kinda crazy that this even needs to be said.

[BeefWithTheChief @HesNotABaby1]

@astrodanish We don't RSI either, as Jeff Ladish pointed out somewhere on here today.