“Rapid capability gain around supergenius level seems probable even without intelligence needing to improve intelligence” by Towards_Keeperhood, Davanchama

Speaker 00s - 516.88s

Rapid capability gain around supergenious level seems probable even without intelligence needing to improve intelligence. By Towards Keeperhood and Davenge Armour TLDR 1. Around Einstein level, relatively small changes in intelligence can lead to large changes in what one is capable to accomplish. A, for example, Einstein was a bit better than the other best fizzy at seeing deep connections and reasoning, but was able to accomplish much more in terms of impressive scientific output.2. There are architectures where small changes can have significant effects on intelligence. A, for example, small changes in human brain hyperparameters, Einstein's brain didn't need to be trained on 3x the compute than normal physics professors for him to become much better at forming deep understanding, even without intelligence improving intelligence. Heading Einstein and the heavy tale of human intelligence. 1905 is often described as the anis mirabulous of Albert Einstein. He founded quantum physicsby postulating the existence of light, Quanta, explained Brownian motion, introduced the special relativity theory and derived E equals MC squared from it. All of this. In one year. While having a full-time job in the Swiss patent office, with the exception of John von Neumann, would say those discoveries alone seem more than any other scientist of the 20th century achieved in their lifetime, though it's debatable. Though perhaps even more impressive is that Einstein was able to derive general relativity.Einstein was often so far ahead of his time that even years after he published his theories the majority of physicists rejected them because they couldn't understand them, sometimes even though there was experimental evidence favouring Einstein's theories. After solving the greatest open physics problems at the time in 1905, he continued working in the patent office until 1908, since the universities were too slow on the uptake to hire him earlier. Subheading. Example for how far ahead of his time Einstein was. Deriving the theory of light quanta.The following section is based on parts of the eighth chapter of surfaces and essences by Douglas Hofstartter. For an analysis of some of Einstein's discoveries, which show how far ahead of his time he was, I can recommend reading it. At the time, one of the biggest problems in physics was that black body spectrum, which describes the spectrum of electromagnetic wavelengthsemitted by a black body. The problem with it was that the emitted spectrum was not explainable by known physics. Einstein achieved a breakthrough by considering light not just as a wave, but also as light quanta. Although this idea sufficiently explained the black body spectrum, physicists, at least almost, unanimously rejected it. The fight between the light is corpuscles and light is a wave faction had been decided a century ago, with a clear victory for the wave faction. Being aware of these possible doubts, Einstein proposed three experiments to prove his idea,one of which was the photoelectric effect. In the following years, Robert Milliken carried out various experiments on the photoelectric effect, which all confirmed Einstein's predictions. Still, Milliken insisted that the light quanta theory had no theoretical basis and even falsely claimed that Einstein himself did not believe in his idea anymore. From surfaces and essences, page 611. To add insult to injury, although the 1921 Nobel Prize in Physics was awarded to Albert Einstein,it was not for his theory of light quanta, but for his discovery of the law of the photoelectric effect. Weirdly, in the citation, there was no mention of the ideas behind that law, since no one on the Nobel Committee, or in all of physics, believed in them. And thus Albert Einstein's revolutionary ideas on the nature of light, that most fundamental and all-pervading of natural phenomena, were not what won him the only Nobel Prize that he would ever receive. Instead, it was just hislittle equation concerning the infinitely less significant photoelectric effect. It's as if the highly discriminating guide Michelin, in awarding its tip-type rank of three stars to Albert's Obarge had systematically ignored its chefs consistently marvellous five-course meals and had cited merely the fact that the Obarge serves very fine coffee afterwards. End quote. Subheading. Concluding thoughts on Einstein.Einstein was able to reason through very complex arguments he constructed via thought experiments without making a mistake. He was able to generalize extremely well from other physics discoveries to get a sense of the underlying nature of physical law. I believe that what enabled Einstein to make key discoveries much faster than the whole remaining field of theoretical physics combined, which itself contained many of the smartest people at the time, was that he was smarter in some dimensions of intelligence than all other20th century scientists, rather than him just being born with good physics particular intuitions. Heading. Takeaways 1. Capabilities are likely to cascade once you get to Einstein-level intelligence, not just because an AI will likely be able to form a good understanding of how it works and use this to optimize itself to become smarter, but also because it empirically seemsto be the case that when you're slightly better than all other humans at stuff like seeing deep connections between phenomena, this can enable you to solve hard tasks like particular research problems much, much faster, as the example. Of Einstein suggests A-A-A-K-A. Around Einstein level, relatively small changes in intelligence can lead to large changes in what one is capable to accomplish. 2. For human brains, small changes in hyperparameters can lead to very significant increases in intelligence. Intuitively, one would suspect that scaling up training compute by 2x is a significantly larger change than than having a plus 6 point for STD hyperparameter sample instead of a plus 5 point for STD1, even though it is notobvious to me that 2X training compute would get you from great physics professor to Einstein if we had transformer architectures. So either there is some grocking cascade around genius level intelligence where capabilities can quickly be learned and improved, or it's just that, human, brain scale significantly faster in performance than transformers currently seem to. A. AKA. For at least some architectures, around genius level, small changes in hyperparameters,or perhaps also compute, can lead to relatively large changes in intelligence. 3. Compute-based AI capability forecasting is unlikely to work well, since this entirely neglects the significant intelligence gap between Einstein and average humans. Heading Requests to AI researchers. Nobody currently knows how to align strongly superhumanly smart AIs to human interests, and we need way more time to solve this problem.Making incremental progress on AI capabilities is shortening the timeline. We have left to figure out how to align AI and is thus making human extinction more likely. Thus by far the best action is to stop advancing AI capabilities. Absent this, please be aware that capabilities might rapidly cascade around genius or super genius level intelligence and take measures accordingly. In particular, one, monitor how quickly performance of an AI is improvingin training. Two, when capability is performing unusually quickly. Stop an audit. A, do not ignore warning signs. If warning signs show up, stop training and coordinate with governments and other AI labs to get more time to solve the alignment problem. B's if the audit is fine, scale up slowly and continue to carefully audit unusual training dynamics. 3. It generally perform regular and precise safety audits while scaling up. 4. Be especially careful when scaling up new architectures or training setups. There likely exist architectures which scale much faster than transformers and might reach superhumanintelligence without needing nearly as much compute as the current best models. This article was narrated by Type 3 audio for less wrong. It was first published on May 6, 2024. The original text contained six footnotes which were omitted from the narration. To report an issue or give feedback on this narration, go to t3a.is.