The most prestigious law school admissions discussion board in the world.
|
 |
 
|
Leopold Aschenbrenner essay on the AI arms race that will change world by 2027
Poast new message in this thread
Date: June 12th, 2024 3:47 PM Author: French exciting son of senegal Really good read and a podcast he did about it. Dude is very smart.
tl;dr version: superintelligence is within reach and the capex to get there is being spent, the teams are being assembled, the power plants built, etc.; eventually US government will step in to oversee the project as a kind of Manhattan Project because a six-month head start is the equivalent of 20-30 years in arms development; things get really weird from a game theory perspective once superintelligence is achieved but before the robots take over. It's US v. China with the Middle East and Russia posing major risks because of their abundant natural energy resources.
Essay: https://situational-awareness.ai/ Four-hour podcast: https://www.dwarkeshpatel.com/p/leopold-aschenbrenner |
Date: June 12th, 2024 4:28 PM Author: French exciting son of senegal (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47740975) |
|
Date: June 12th, 2024 4:50 PM Author: big giraffe stage pretty chill that the world will end in five years and I spend my precious remaining days poasting on a decrepit law school admissions forum (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47741056) |
 |
Date: June 12th, 2024 6:14 PM Author: French exciting son of senegal It’s part of the data set (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47741277) |
|
Date: June 12th, 2024 7:22 PM Author: Adventurous walnut jewess (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47741556) |
|
Date: June 12th, 2024 7:25 PM Author: Citrine bawdyhouse *superintelligent AI addressing the world* "I'm gay and retarded" |
|
Date: June 12th, 2024 7:26 PM Author: Adventurous walnut jewess 180000000000000000000000000000 (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47741570) |
|
Date: June 14th, 2024 9:03 PM Author: French exciting son of senegal (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47748063) |
|
Date: June 12th, 2024 7:22 PM Author: twinkling orchestra pit He just had another podcast with Chollet that was interesting and raised some concerns about this LLM maximalist point of view. LLMs still perform quite poorly on ARC. There are real concerns that transformer scaling will not address all the problems with out of distribution generalization and algorithmic reasoning. This buys a bit of time and probably pushes out timelines past 2027. Things will become progressively weirder though over the next decade and it’s hard to see how AGI is out more than 15 years. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47741554) |
|
Date: June 12th, 2024 7:29 PM Author: Irate Location nothing being done in "ai" right now is a road to actual intelligence. what we have right now is essentially a sophisticated version of a database that can accept unstructured queries and that can interpolate between dataset entries. it would take a complete paradigm shift to get to something qualitatively more powerful than that. may or may not happen. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47741579) |
|
Date: June 13th, 2024 11:34 AM Author: French exciting son of senegal This is not right. The trend lines are clear. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47743102) |
|
Date: June 13th, 2024 12:51 PM Author: Irate Location no (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47743420) |
|
Date: June 13th, 2024 1:23 PM Author: twinkling orchestra pit Most people seem like pattern matchers who aren’t able to robustly deal with novel situations. I would trust GPT-4 more to solve a random math word problem or answer RC questions from an LSAT than most people. The interpolation/extrapolation distinction becomes less and less meaningful in high dimensional spaces when you are training on everything. I agree LLMs don’t go all the way but they very likely go a lot further than they do now and there are other things in the AI space used for game playing that look like they can be combined with LLMs to bootstrap capabilities. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47743513) |
|
Date: June 13th, 2024 2:01 PM Author: Irate Location llms far surpass humans at defined, stereotyped tasks that we have a lot of data for. they can't generalize to unseen environments (new probability distributions) and they lack an inherent teleology to prompt them interact with those environments. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47743680) |
|
Date: June 13th, 2024 2:47 PM Author: twinkling orchestra pit I agree with this but most people suck at dealing with novel situations too and they can’t even do the routine cognitive tasks that GPT-4 excels at. I don’t think a median human could be made competitive with it at reading comprehension tasks. With 100x the training compute, there is no chance. There is really no principled reason why LLMs paired with model based RL couldn’t be made data efficient and adaptable. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47743890) |
|
Date: June 13th, 2024 4:14 PM Author: Irate Location this is silly (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47744229) |
|
Date: June 13th, 2024 11:39 AM Author: Cerise tripping queen of the night I, for one, cant wait for AI to do all my TPS reports for me. Whatever will I do once those are automated? Never mind that basically everything we do in Compliance is bespoke transaction monitoring and the vanilla bulk transaction stuff was automated 24-25 years ago and hasnt had humans involved in a generation anyway.
Like the biggest thing is Compliance now is marijuana and the hemp exemption BS that plagues us. The fuck is AI going to do about any of that? |
|
Date: June 13th, 2024 11:43 AM Author: French exciting son of senegal The interesting thing about his essay is the geopolitics part. The time when AGI is pursued by private companies is coming to a close. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47743136) |
|
Date: June 16th, 2024 7:28 PM Author: French exciting son of senegal (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#47752081) |
|
Date: October 27th, 2024 3:00 PM Author: twinkling orchestra pit Ivanka read it and endorsed it. Can’t believe how much AI is starting to affect elite thinking
https://x.com/IvankaTrump/status/1839002887600370145 |
|
Date: October 27th, 2024 5:15 PM Author: French exciting son of senegal There’s like a five percent chance of total destruction and a five percent chance of positive total transformation. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#48247027) |
|
Date: October 27th, 2024 6:20 PM Author: twinkling orchestra pit And the other scenarios? I don’t see much room for an outcome in between. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#48247316) |
|
Date: October 27th, 2024 6:22 PM Author: Domesticated big-titted twinkling uncleanness pretty sure something like Idiocracy covers most of the 90% (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#48247323) |
|
Date: October 27th, 2024 6:24 PM Author: Nudist cruise ship Nope. It's going to be a post-scarcity utopia, extinction, or some sort of truly weird outcome like everyone being uploaded into a simulation of something retarded for eternity. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#48247333) |
|
Date: October 27th, 2024 8:36 PM Author: twinkling orchestra pit Right. Like what would be a plausible middle ground with hyper intelligent machines? You would have to believe intelligence tops at exactly the human level, when we have many reasons for believing that is not the case. Life can not go on in a normal way. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#48247755) |
|
Date: November 1st, 2024 9:37 PM Author: French exciting son of senegal (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#48270046) |
|
Date: July 31st, 2026 12:16 AM Author: .,.,.,.,,,,.,,.,.,.,...,,..,, Where r we on this (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#50035998) |
|
Date: July 31st, 2026 12:23 AM Author: The Penis We are currently at a point where AI has surpassed almost all if not all humans in intelligence. And it's not even 2027 yet. (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#50036002) |
|
Date: July 31st, 2026 12:25 AM Author: .,.,.,.,,,,.,,.,.,.,...,,..,, lol we’ve got a live one here folks (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#50036007) |
|
Date: July 31st, 2026 12:26 AM Author: The Penis Do you think you are more intelligent than Claude Fable? (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#50036009) |
|
Date: July 31st, 2026 12:45 AM Author: .,.,.,.,,,,.,,.,.,.,...,,..,, Yes (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#50036028) |
|
Date: July 31st, 2026 4:46 PM Author: valar vettiri morghulis At what? (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#50037372) |
|
Date: July 31st, 2026 1:57 AM Author: Ceci n'est pas un pumo He seems extremely gay (but I guess he isn't?)
https://x.com/dnapway/status/2082876150322540657 |
|
Date: July 31st, 2026 4:46 PM Author: valar vettiri morghulis That’s what Alpha zoomers look and sound like now (http://www.autoadmit.com/thread.php?thread_id=5539938&forum_id=2/en-en/#50037373) |
|
Date: July 31st, 2026 4:46 PM Author: valar vettiri morghulis lol at this super genius being a) right and b) blowing up his fund with leverage and c) having to sell all his positions to Citadel.
|