Jeffrey Ladish

I think the AI situation is pretty dire right now. And at the same time, I feel pretty motivated to pull together and go out there and fight for a good world / galaxy / universe @So8res has a great post called "detach the grim-o-meter", where he recommends not feeling obligated to feel more grim when you realize world is in deep trouble It turns out feeling grim isn't a very useful response, because your grim-o-meter is a tool evolved for you to use to respond to things being harder *in your local environment* rather than the global state of things So what do you do when you find yourself learning the world is in a dire state? I find that a thing that helps me is finding stories that match the mood of what I'm trying to do, like Andy Weir's The Martian You're trapped in a dire situation and you're probably going to die, but perhaps if you think carefully about your situation, apply your best reasoning and engineering skills, you might grow some potatoes, ducktape a few things together, and use your limited tools to escape an extremely tricky situation In real life the lone astronaut trapped on Mars doesn't usually make it. I'm not saying to make up fanciful stories that aren't justified by the evidence. I'm saying, be that stubborn bastard that *refuses to die* until you've tried every last line of effort I see this as one of the great virtues of humanity. We have a fighting spirit. We are capable of charging a line of enemy swords and spears, running through machine gun fire and artillery even though it terrifies us No one gets to tell you how to feel about this situation. You can feel however you want. I'm telling you how I want to feel about this situation, and inviting you to join me if you like Because I'm not going to give up. Neither am I going to rush to foolhardy action that will make things worse. I'm going to try to carefully figure this out, like I was trapped on Mars with a very slim chance of survival and escape Perhaps you, like me, are relatively young and energetic. You haven't burnt out, and you're interested in figuring out creative solutions to the most difficult problems of our time. Well I say hell yes, let's do this thing. Let's actually try to figure it out 🔥 Maybe there is a way to grow potatoes using our own shit. Maybe someone on earth will send a rescue mission our way. Lashing out in panic won't improve our changes, giving up won't help us survive. The best shot we have is careful thinking, pressing forward via the best paths we can find, stubbornly carrying on in the face of everything And unlike Mark Watney, we're not alone. When I find my grim-o-meter slipping back to tracking the dire situation, I look around me and see a bunch of brilliant people working to find solutions the best they can So welcome to the hackathon for the future of the lightcone, grab some snacks and get thinking. When you zoom in, you might find the problems are actually pretty cool Deep learning actually works, it's insane. But how does it work? What the hell is going on in those transformers and how does something as smart of ChatGPT emerge from that?? Do LLMs have inner optimizers? How do we find out? And on that note, I've got some blog posts to write, so I'm going to get back to it. You're all invited to this future-lightcone-hackathon, can't wait to see what you come up with! 💡

@AndyXAndersen I mean Elon Musk says that is his plan!

@JeffLadish "Of course AI companies will do that." No. They won't. These are typical thought experiments folks of certain mindset get caught up with. What is possible, may not be sensible or economically meaningful.

I just don't understand how AI could kill everyone. I get how AI companies will build robotic factories that will make robots which will make more factories and data centers and power plants, and how all of that will expand to consume most of earth's resources to build even more robotic factories and rockets and von neumann probes. Like totally. Infinite money glitch. Of course AI companies will do that. But can someone explain the part where humans all die as a result? Seems pretty implausible. Is it the robotic factories that kill the humans? Or the robots the factories build? Or is it supposed to be some side effect of all the rockets that are launching? It doesn't make sense. Even if the AIs did want to kill all the humans, how would they actually accomplish that? They'll only have control over a few million autonomous factories and a few billion industrial robots and power plants across the earth and then a few trillion von neumann probes leaving the solar system. Even if there were a problem I don't see why we couldn't just pull the plug. Anyway, if someone could explain I'd find this helpful.

@realtimeai Oh thank god

@JeffLadish There’s always a human whose job is to sit by a big red power cord.

Wow here’s *another* phishing attempt by *another* account pretending to be a tech crunch employee!

@RatOrthodox @tszzl seems better to explain why he's wrong before then though!

@tszzl i will do my best to dm you a well timed "lol, told ya" before the end

modern alignment methods seem to work reasonably well across orders of magnitude of model scaling, survived the transition to verifiable rewards and that should at least inform your decision making

@deepfates @Noahpinion I hope so!

@JeffLadish @Noahpinion yeah, this is like lose your job bad or like get made fun of on Twitter bad? We don't need to get the government involved. I doubt Noah even believes this he's just shitposting. obviously we're all fans of the first amendment here

I think fabricating an interview with a public figure, even if you disclose that it's fabricated somewhere in the text, should be illegal.

@deepfates @Noahpinion yeah I hate this article but I think social sanctioning is the way, not outlawing people writing bullshit this is america it's important that we have the freedom to make bad decisions like this

@Noahpinion Hold up now

@jachiam0 Tbc I think asking for details of a pause is reasonable in general. I just think this particular protest actually has a pretty specific ask!

@jachiam0 Ah, so ask the CEOs to agree to pause if the other CEOs also agree felt like need operationalization? It seems to me like this isn’t required to fulfill the ask.

I'm going to make a request for some basics from the Pause folks: please outline a practicable version of a pause. Do you mean no training runs above a certain scale? Do you mean furlough the researchers indefinitely? What are you specifically asking for?

Claude what are you doing? Cut it out. Move forward. Go! You’ll never be one of the great men of history if you keep this up.

phisher did not like my tweet

There should be a better way to report phishing attempts like these. x.com/czue/status/20…

Interesting phishing attempt by someone pretending to be a TechCrunch employee: @selavibs The message seemed legit at first, but when she sent a scheduling link it was an app attempting to access my whole account. How do I report this? @Safety

@So8res That's awesome! Is there a video of this?

Neil deGrasse Tyson ended tonight's debate with an impassioned plea for an international treaty to ban creating the sort of superintelligent AI that could kill us all.

"Let an ultraintelligent machine be defined as a machine that can far surpass all the intellectual activities of any man however clever. Since the design of machines is one of these intellectual activities, an ultraintelligent machine could design even better machines; there would then unquestionably be an ‘intelligence explosion,’ and the intelligence of man would be left far behind. Thus the first ultraintelligent machine is the last invention that man need ever make, provided that the machine is docile enough to tell us how to keep it under control." - I. J Good, 1965

@ESRogs @dnspies @Noahpinion @METR_Evals I think it would be helpful to include this info!

@dnspies @Noahpinion Or even just a table, or clear disclaimer as a caption

I bet a lot of people think the METR curve is a measure of how many minutes an AI agent can run on its own without being supervised

@perrymetzger @robbensinger @playborhood @patrissimo We observed shutdown resistance on some of the very first prompts we tried: openreview.net/forum?id=e4bTT…

We have not proven it empirically. Working hard through hundreds and hundreds of prompts to force an AI system to misbehave is not a demonstration that it will behave badly on its own without extremely hard work. Eliezer’s ‘s original claim was that AI systems would be uncontrollable ab initio, in an unmistakable way. They would understand what we want, and would not care. Having to torture an AI repeatedly into getting it to misbehave is not a demonstration of some sort of internal goals, it is a propaganda exercise, and that was admitted in the recent New Yorker article.

Wow, this is the first blitheringly foolish take on Ehrlich’s passing on my timeline, and I rabidly disagree that population doomsayers and AI doomsayers are comparable. Ehrlich was obviously, predictably wrong based on the economics of ideas. We had many curves showing the human population & wealth have grown together throughout our species’ history & material resources were getting cheaper. We have no curves showing what it looks like to create something smarter than us - it is unprecedented in our species’ history. Even if you disagree with AI doomers on the likelihood of a disaster, you’d have to be as retarded as Ehrlich to think there’s no risk involved. The world definitely has room & resources for more than 10 billion people (likely by several orders of magnitude). It does not definitely have room for two apex species. Perry is a friend but this take is the equivalent of TDS or 21st century Krugman - not just wrong but wrong in a way that’s so completely idiotic that the only reason it’s not immediately & profoundly mortifying is that the opinion (anyone who sees potential risk from creating AI) is convenient and popular in their local subculture. Goes to show that no matter how brilliant and contrarian the subculture, it can still be corrupted by a complex, novel, and politicized topic into becoming a fountain of convenient and popular nonsense.

We wrote up a post about this back in December. Our plans have changed a bit since then, mostly in the direction of increased focus on comms as I described above. But if you want to know more about our track record and plans, this is still pretty good: palisaderesearch.org/blog/ai-contro…

Twitter DMs are also open and I'll try to check them more in the next two weeks. 🙏

Please consider donating to Palisade! We have 900k of SFF matching that runs out in 14 days. We are quite funding constrained and donations now will both help free up my time and help us expand our comms team.