TL;DR: There are many posts on the Alignment Forum/LessWrong that could easily be on arXiv. Putting them on arXiv has several large benefits and (sometimes) very low costs. … | Continue reading
I keep finding cause to discuss the problem of the criterion, so I figured I'd try my hand at writing up a post explaining it. I don't have a great track record on writing clear explanations, but I'l… | Continue reading
Lee Smolin's book The Trouble with Physics: The Rise of String Theory, the Fall of a Science, and What Comes Next is ostensibly about why string theory can't solve what he calls the Five Great Proble… | Continue reading
In the first half of the 14th century, the Franciscan friar and logician, William of Occam proposed a heuristic for deciding between alternative explanations of physical observables. As William put i… | Continue reading
One day, we might be able to bring back to life every human ever lived, by the means of science and technology.And it will be a good day. … | Continue reading
WHAT ARE THE RESTRICTIONS?Broadly speaking, the policy restricts the sale of GPUs and related technology to Chinese companies. The specifics remain unclear because (a) the Department of Commerce has… | Continue reading
You know those health books with “miracle cure” in the subtitle? The ones that always start with a preface about a particular patient who was completely hopeless until they tried the supplement/medit… | Continue reading
Unless I'm eating with other people, food for me is fuel. … | Continue reading
After doing a bit of digging, I'm pretty convinced that nuclear famine was never a thing to begin with. Maybe back during the Cold War, when the Soviets were seriously considering ground-bursting tho… | Continue reading
A community blog devoted to refining the art of rationality | Continue reading
A community blog devoted to refining the art of rationality | Continue reading
On occasion, for my work at Lightcone I have been able to buy things faster than their advertised lead times. For example, I once got… … | Continue reading
A 2006 study showed that “280,000 people in the U.S. receive a motor vehicle induced traumatic brain injury every year” so you would think that wearing a helmet while driving would be commonplace. Ra… | Continue reading
A community blog devoted to refining the art of rationality | Continue reading
A community blog devoted to refining the art of rationality | Continue reading
What explains the hockey-stick shape of world GDP over time, with seemingly no progress for thousands of years, followed by soaring growth? … | Continue reading
We had some discussions of theAGI ruin arguments within the DeepMind alignment team to clarify for ourselves which of these arguments we are most concerned about and what the implications are for our… | Continue reading
First, a few paper titles: • * Pretrained Transformers as Universal Computation Engines * Can Wikipedia Help Offline Reinforcement Learning? (Answer: yes.) * Pretrained Transformers Improve Out-of… | Continue reading
Note: As usual, Rob Bensinger helped me with editing. I recently discussed this model with Alex Lintz, who might soon post his own take on it. • … | Continue reading
List of blogs from the diaspora and rationalist movement. Last large update: 2017-05-25GENERAL RATIONALITY AND PHILOSOPHYAn Algorithmic Lucidity Zack M DavisCompass Rose Ben Hoffman (Benquo)Don’t Worry About the Vase Zvi MoshowitzEntirely UselessEverything Studies John NerstGrogn … | Continue reading
This article is a writeup of the conversation at a meetup hosted by Austin Less Wrong on Saturday, February 27, 2021. The topic was the winter weather and infrastructure crisis that took place the pr… | Continue reading
In 2020, proposed scaling laws for language model seemed to indicate data wasn't as important as model parameters to final performance. The optimal distribution of resources was something like 5:2, f… | Continue reading
I think thatin the coming 15-30 years, the world could plausibly develop “transformative AI”: AI powerful enough to bring us into a new, qualitatively different future, viaan explosion in science and… | Continue reading
Yann LeCun recently posted A Path Towards Autonomous Machine Intelligence, a high-level description of the architecture he considers most promising to advance AI capabilities. … | Continue reading
A community blog devoted to refining the art of rationality | Continue reading
I havebeen convinced tobelieve that looking at the gap between human inner and outer alignment is a good way to think about potential inner/outer alignment problems in artificial general intelligence… | Continue reading
INTRODUCTIONWhen I first heard about EfficientZero, I was amazed that it could learn at a sample efficiency comparable to humans. What's more, it was doing it without the gigantic amount of pre-trai… | Continue reading
Last August, my research group created a forecasting contest to predict AI progress on four benchmarks. Forecasts were asked to predict state-of-the-art performance (SOTA) on each benchmark for June… | Continue reading
TL;DR: We’re launching theInverse Scaling Prize: a contest with $250k in prizes for finding zero/few-shot text tasks where larger language models show increasingly undesirable behavior (“inverse scal… | Continue reading
Since I've written the article on structured concurrency and implemented libdill the progress went on. … | Continue reading
EXECUTIVE SUMMARYUsing a dataset of 470 models of graphics processing units (GPUs) released between 2006 and 2021, we find that the amount of floating-point operations/second per $ (hereafter FLOP/s… | Continue reading
DeepMind: The Podcast - Season 2 was released over the last ~1-2 months. The two episodes most relevant to AGI are: … | Continue reading
I've argued that the development of advanced AI could make this the most important century for humanity. A common reaction to this idea is one laid out by Tyler Cowen here: "how good were past thinke… | Continue reading
A Paperclip Maximizer is a hypothetical artificial intelligence whose utility function values something that humans would consider almost worthless, like maximizing the number of paperclips in the universe. The paperclip maximizer is the canonical thought experiment showing how a … | Continue reading
A Chemical Hunger (a), a series by the authors of the blog Slime Mold Time Mold (SMTM) that has beenreceived positivelyon LessWrong, argues that the obesity epidemic is entirely caused (a) by environ… | Continue reading
BACKGROUNDI have been doing red team, blue team (offensive, defensive) computer security for a living since September 2000. The goal of this post is to compile a list of general principles I've lear… | Continue reading
Last year, GitHub announced their Copilot system, an AI assistant for developers based on OpenAI's Codex model, as a free closed beta. Yesterday, they added that Copilot would now be available to eve… | Continue reading
At the Singularity Summit 2007, one of the speakers called for democratic, multinational development of artificial intelligence. So I stepped up to the microphone and asked: … | Continue reading
Douglas Hofstadter wrote in the Economist recently on whether modern large language models are conscious. As part of this, he and his colleague David Bender claim that GPT-3 has a "mind-boggling holl… | Continue reading
As was said before: "A lot of the AI risk arguments seem to come... with a very particular transhumanist aesthetic about the future (nanotech, ... etc.). I find these things (especially the transhuma… | Continue reading
PREAMBLE:(If you're already familiar with all basics and don't want any preamble, skip ahead toSection B for technical difficulties of alignment proper.) … | Continue reading
This is part of a weekly reading group onNick Bostrom's book,Superintelligence. For more information about the group, and an index of posts so far see theannouncement post. For the schedule of future… | Continue reading
People seem to be continually surprised, over and over again, by the new capabilities of big machine learning models, such as PaLM, DALL-E, Chinchilla, SayCan, Socratic Models, Flamingo, and Gato (al… | Continue reading
Note: This is a joint distillation of bothIterated Distillation and Amplification by Ajeya Cotra (summarizing Paul Christiano) andA Generalist Agent by DeepMind. … | Continue reading
I. THE SEARCH FOR DISCONTINUITIESWe’ve been looking for historic cases of discontinuously fast technological progress, to help with reasoning about the likelihood and consequences of abrupt progress… | Continue reading
I got access to DALL-E 2 earlier this week, and have spent the last few days (probably adding up to dozens of hours) playing with it, with the goal of mapping out its performance in various areas – a… | Continue reading
Why are costs of certain things, most notably education and healthcare, skyrocketing so quickly, with relatively little improvement in quality? A few years ago, SlateStarCodex and Marginal Revolution… | Continue reading
The long term futuremay be absurdand difficult to predict in particulars, but much can happen in the short term. … | Continue reading