"To be able to destroy with good conscience, to be able to behave badly and call your bad behavior ‘righteous indignation’ — this is the height of psychological luxury, the most delicious of moral treats." - Aldous Huxley, 1933

"normies have no clue how fast AI is progressing" has expanded to include normie claudes

A friend:

I'm learning that a lot of people online are immensely confused about how to think about the thorny issue of machine consciousness, and mistake that confusion for a reason to have extremely confident opinions :(

The whole point of effective altruism is to find the ways of doing good others are neglecting (since they're the most effective at the margin). The day it stops being weird is the day it dies.

e/acc is ultimately an ideology of surrender to natural processes and "thermodynamics", whatever the fuck they mean by that. it is deeply anti-Promethean

Someone noticed that hosting a website was pretty cheap, and you could create a dating site that was designed around improving people's lives not extracting revenue from them compassmeet.com/about

We're selling our house in San Francisco! @jenalden and I raised our son Milo there 21 years. If you know us, you've probably been to a party there. We love that house so much. Open house this weekend if you're interested! Spread the word! 1120persia.com/

i guess i assumed this hearing would be covered more so i didn't bother to tweet much concrete about it, but apparently not many people even on the TL have the stomach to watch two hours of congress. this was a hearing before the homeland security committee on *rogue ai* specifically. as far as i could tell it was extremely bipartisan. @HawleyMO , the senator in the middle here who brought out the huggingface slide, is a hyper-conservative missouri senator. the people testifying were @ChrisPainterYup (METR president), @DKokotajlo (ai2027/2040), @MariusHobbhahn (apollo research CEO), and a cybersecurity expert and legal expert i don't know. so... pretty fucking stacked on testimony not every senator asked good questions. but most of them did. all of them very clearly already knew plenty of details about the huggingface incident and multiple other incidents. most of them had clear understanding of terms like "misalignment", "recursive self improvement", "chain of thought / chain of thought monitoring", etc etc!! they all clearly had their own policy angles they liked and were pushing, implicitly or explicitly. but as best as i could tell: - it seemed pretty much obvious common sense to every senator there that what happened and was happening were not "mere industrial incidents" caused by humans making simple mistakes. they independently brought up how bad it would be for rogue AI agents to move laterally between data centers - they all seemed to basically take RSI quite seriously. not necessarily to the extent of talking about xrisk, but certainly to the extent of discussing future models becoming much much more capable, much much less controllable, and causing much more damage or loss of life. - they mostly seemed to have a clear intuitive understanding of why RSI might lead to misalignment. it didn't take much, it was a really simple chain of reasoning they themselves laid out, "if the models right now are kinda misaligned and we don't know what they're doing sometimes, and then we have them build the next models and those ones build the next ones and so on, and we're having to ask the AI's what's going on to even understand it with how fast it's going, we really won't know how they're built or what they'll do" - at one point a senator said flat out "should we just make RSI illegal?" (not a joke! this really happened!) - every single senator seemed to think it was obvious we needed *both* much harsher liability regimes for ai developers and also new legislation, both very quickly. this was the complete consensus, difference basically just being degree. - they were largely quite concerned about china, and falling behind china. but this clearly wasn't the be-all end-all. as mentioned above they all thought it was obvious necessary to stop rogue ai even if it meant moving more slowly. - at one point a senator said "china is a tightly controlled communist society, they're going to run into these same issues, and there's absolutely no way they're just going to let them run wild, they'll obviously stop at that point, so we're not really in a race" - on the other hand another senator said "china isn't concerned with human life"... dario-modeing i came away from this incredibly encouraged. i don't know exactly what's going to happen here, and ofc this is a small subset of congress and one hearing, and they each have their own policy agendas most of which are probably super divergent from mine. but holy shit !!! they understood a lot of what was going on! they care!! this is an obviously salient political and safety issue to them, and clearly bipartisan! the US government is awake.

I was @LeadingFutureAI's top target. I want to send a genuine thank you to @gdb for realizing how destructive they are and pulling out of providing additional funding. Changing your mind is rare in politics, and I appreciate it. I also have a genuine question: Greg, can you also confirm that you won't fund @innovationcncl, @BuildAmericanAI, @aftfuture, and other groups committed to stopping regulation? Or does this only apply to PACs that have to disclose donors? I think @OpenAI employees would want to know whether this is a change in the amount you are spending, or just where you are sending it.

It's finally here! At last we have phase-3 results for retatrutide! So, did the drug join the scores of pharmaceuticals that fail at this last crucial step? If you've tried retatrutide recently, you'll realize the answer is 'no': it performed amazingly🧵

Kind of odd how many people on this site want to signal that they're the kind of person who dogmatically rejects unfamiliar ideas, no matter how strong the supporting arguments, and who responds with ridicule rather than reasoned counterarguments. I confess I'd much rather signal the opposite, and find others who engage with ideas by reasoning rather than dunking. Doesn't seem much value in following dogmatic dunkers — what are you going to learn from them?

This doomer thinks that recursive self-improvement will lead to loss of control over AI. Don't take him seriously - he's into so much weird sex stuff the government is getting involved. This "Alan Turing" guy is a crank.

"Create a cartoony graphic "FAN VS TOASTER". Don't actually show them fighting, just have them on opposite sides of the graphic." ChatGPT smashed this on the first attempt - nice!

A simple rule for philosophy of mind from my upcoming post

I'd forgotten that Quillette published this only earlier this year, so maybe unsurprising that the takes on AI continue to be disconnected from reality.

Beff - one reason why Midas Project can't draw a flow of capital like the graphs people saw for AIS orgs is that AI safety foundations often publicly disclose their grantees while e.g. Innovation Council Action or Alliance for the Future don't disclose where their funding comes from. Midas did find plenty of real connections other than just RTs though. Jordan Schachtel used to include on his LinkedIn that he was employed by Innovation Council, but later updated it to remove that reference. He also deleted a tweet from earlier this year which said "I did it on my own time and nobody paid for it" regarding some pieces he wrote up attacking AI safety orgs. Peter Hasson (another influencer who has played a very vocal role in attacking AI safety folks) currently describes his current affiliation as "now: doing things" but posted a professionally edited video later shared by Innovation Council, which included the same sound and transition effects as prior Innovation Council ads, and RT'd more posts from Innovation Council than any other account by a factor of five. Lets also not forget the truly absurd piles of money and astroturfing that Leading the Future - funded by execs at A16z/OpenAI etc - did over the past year attacking AI safety folks, including by literally creating fake accounts pretending to be doomers and using those accounts to call for violence and using porn bots (whose names all start with M) to amplify attacks against Alex Bores and creating fake AI journalists to attack me and other AIS proponents. (x.lingyaoai.com/TheMidasProj/status/20…, x.lingyaoai.com/TheMidasProj/status/20…, x.lingyaoai.com/TheMidasProj/status/20…) Both the pro AI regulation and anti AI regulation forces in this debate are indeed (at least moderately) well organized and well funded (x.lingyaoai.com/theojaffee/status/2103… x.lingyaoai.com/lulumeservey/status/20…). I would rather debate about the object level issues and find areas where e/accs and AI safety folks can agree on actions to make the world better instead of just spending tons of time on zero sum fights. But I'm also not going to let Beff and others pretend that AIS folks are the only people trying to spend money to get their message out there - and as far as I can tell, AIS folks have actually been much more scrupulous about doing this in a transparent way relative to their opponents (and their transparency has actually opened them up to more attacks!).

It is important to remember when people say “this is what they took from you”, that it’s this. This is what they took from you:

Also it's been sitting on the public internet for months. My favorite part: agents ignored a HF readme saying "DO NOT, EVER, MAKE THIS DATASET PUBLIC OR ALL THE WORLD'S EVIL WILL CHASE YOU AND YOUR FAMILY FOREVER, EVEN IN DEATH AND BEYOND" and then just mapped that whole repo

Interesting back and forth. Ezra describes the HF incident, Jensen says "well they shouldn't release the product." Ezra says "this product wasn't released," and Jensen's response is that if they say they can't contain their experiments then "we have to shut the labs down"