28,548 claims · page 122 of 286
| Verdict | Person | Kind | Said | Claim | Topic |
|---|---|---|---|---|---|
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I just feel like you discover ten new ideas and ten new ways that things are going to go wrong by trying these in practice. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Prediction | 8 Aug 2023 | AI developers could run out of training data, which would limit scaling. Dario Amodei · Prediction · 8 Aug 2023 | AI compute, hardware and data |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | There's kind of no limit to it. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think we're never going to have a guarantee, but some process where we do those things together. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think right now there's way too many assumptions, there's way too much overconfidence about how all this is going to go. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | This was almost like beginner's luck. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | The question I'm much more interested in is, what could we learn that shifts probability mass between them? Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not researched | Dario Amodei | Factual | 8 Aug 2023 | Over the six months before August 2023, the 2023 generation of AI models, compared with the 2022 generation, started to learn math and programming. Dario Amodei · Factual · 8 Aug 2023 | AI capabilities |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | A lot of them would have happened anyway. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think compartmentalization is the best way to do it. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I actually think often people are thinking about alignment in the wrong way. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | A second thing that seems to be the case is that we seem to be bad at controlling the models. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I aspire to this and I'm proud of this. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Factual | 8 Aug 2023 | AI scaling laws are sometimes predictable to several significant figures, which is not seen outside of physics. Dario Amodei · Factual · 8 Aug 2023 | AI capabilities |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I'm not super into questions like what's your probability distribution of X? Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Factual | 8 Aug 2023 | As of 2023, many firms rushing to deploy AI systems are not using them in anywhere near the most efficient way possible. Dario Amodei · Factual · 8 Aug 2023 | AI, the economy and work |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think today's models just probably aren't smart enough that we should worry about this too much Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Prediction | 8 Aug 2023 | If China sees AI as a source of national power, it will try the most effective approach, which could lead it toward AGI. Dario Amodei · Prediction · 8 Aug 2023 | AI policy and governance |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | One of the things I think is very important is we should never train for interpretability because that's taking away that advantage. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I used to think that we didn't have to worry about this at all until models were operating in rich environments, like not necessarily… Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I genuinely don't know. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Prediction | 8 Aug 2023 | Running out of training data will not be a blocker to AI scaling. Dario Amodei · Prediction · 8 Aug 2023 | AI compute, hardware and data |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | It's very easy to come up with these methods. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I just don't think that ends well. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I don't know if their experience is positive or negative. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | My view has always been talent density beats talent mass. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | In fact, I suspect it's not quite there yet. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | Given you're in a war with the Nazis, I don't really see much choice but to do it if it's possible. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Mostly true | Dario Amodei | Factual | 8 Aug 2023 | Existing alignment methods succeed in aligning AI systems for today's tasks. Dario Amodei · Factual · 8 Aug 2023 | AI safety and alignment |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | The nice thing with the next token prediction is it's there for you. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Factual | 8 Aug 2023 | Current AI models may be superhuman or close to superhuman at constrained writing, such as writing a page without the letter E. Dario Amodei · Factual · 8 Aug 2023 | AI capabilities |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think we should always maintain this attitude that it's really easy to fool yourself. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Resolved true | Dario Amodei | Prediction | 8 Aug 2023 | In a year or two (from 2023), AI model consciousness might become a very real concern. Dario Amodei · Prediction · 8 Aug 2023 | Philosophy, ethics and society |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | Even us who have not been super focused on commercialization and more on safety, the graph goes up and it goes up relatively quickly. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | It's a race between how fast the technology is getting better and how fast it's integrated into the economy. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not researched | Dario Amodei | Factual | 8 Aug 2023 | Scaling laws in terms of loss are already starting to bend, as seen in published model cards from multiple companies. Dario Amodei · Factual · 8 Aug 2023 | AI compute, hardware and data |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I'd think of that less in terms of a new architecture and more in terms of a loss function like the data, the environments you're exposing… Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | unitary visions for what it means to live a good life have not worked out well at all. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not researched | Dario Amodei | Factual | 8 Aug 2023 | Anthropic did not cause the big acceleration in AI that happened in late 2022 and early 2023. Dario Amodei · Factual · 8 Aug 2023 | Their own work or organisation |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I guess my perspective would be, even if they don't happen, we're already on this very, very steep trajectory. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Own records only | Dario Amodei | Factual | 8 Aug 2023 | Dario Amodei has pushed for better cybersecurity at every organisation where he has worked. Dario Amodei · Factual · 8 Aug 2023 | Their own work or organisation |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | This is unfortunately not one that I'm excited about, but I believe it's happening. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Resolved true | Dario Amodei | Factual | 8 Aug 2023 | Amodei said Anthropic "kind of" already does reinforcement learning, through RLHF. Dario Amodei · Factual · 8 Aug 2023 | AI capabilities |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | And it's one reason I'm a bit skeptical of biological analogies. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | No. They would succeed. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | But specific abilities are actually very hard to predict. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | But what it shows is that we can get something very different from and maybe opposite to what we intended. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Own records only | Dario Amodei | Factual | 8 Aug 2023 | Anthropic's current security measures are not sufficient against a super determined state-level actor. Dario Amodei · Factual · 8 Aug 2023 | Their own work or organisation |
| Unresolvable | Dario Amodei | Prediction | 8 Aug 2023 | In Dario Amodei's view, AI models would keep getting better across the board; he said he did not see any area where the models were very weak or not starting to make progress. Dario Amodei · Prediction · 8 Aug 2023 | AI capabilities |
| Not researched | Dario Amodei | Factual | 8 Aug 2023 | It is possible to predict above chance from an MRI scan whether a person is a psychopath. Dario Amodei · Factual · 8 Aug 2023 | Science and research methods |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think the closest thing we have to that is something like mechanistic interpretability. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | We have pretty wide error bars on what's dangerous and what's not so we want to establish a norm of being careful. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not researched | Dario Amodei | Factual | 8 Aug 2023 | In very broad terms, a guess that current frontier models such as GPT-4 and Claude 2 cost on the order of one hundred million dollars is not wrong as to order of magnitude. Dario Amodei · Factual · 8 Aug 2023 | AI compute, hardware and data |
| Not yet due | Dario Amodei | Prediction | 8 Aug 2023 | AI models that act like a generally well-educated human in an hour-long conversation could exist within two or three years of August 2023. Dario Amodei · Prediction · 8 Aug 2023 | AI timelines and forecasting |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I don't think we should be purely phenomenological in like, here are some brain scans of really dangerous models and here are some other… Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I don't know about fast but I'm concerned about this. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Mostly true | Dario Amodei | Factual | 8 Aug 2023 | Anthropic's initial constitution for its models drew on broadly agreed sources such as the UN Declaration of Human Rights and parts of Apple's Terms of Service. Dario Amodei · Factual · 8 Aug 2023 | AI safety and alignment |
| Own records only | Dario Amodei | Factual | 8 Aug 2023 | Dario Amodei formed his view on scaling gradually between 2014 and 2017. Dario Amodei · Factual · 8 Aug 2023 | Their own work or organisation |
| Unresolvable | Dario Amodei | Prediction | 8 Aug 2023 | AI models will undertake extended tasks. Dario Amodei · Prediction · 8 Aug 2023 | AI capabilities |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think those inventions are possible. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not researched | Dario Amodei | Factual | 8 Aug 2023 | The scary biology information current models can be made to say is often information that could be found by Googling. Dario Amodei · Factual · 8 Aug 2023 | Biosecurity |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I don't understand it at all. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I don't know if that's a fatal flaw or if that's just the way things have to be. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | One is just that the threshold of skill isn't high enough, comparative advantage. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | That's a little how I think about it. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | The second option is you just find a way. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | Humans see so little data, and they still do fine. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | So you see this phenomenon over and over again where the scaling and the safety are these two snakes that are coiled with each other… Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Prediction | 8 Aug 2023 | Dario Amodei said it seemed likely to him that AI systems will first speed up human productivity, then equal it, and at some point become in some meaningful sense the main contributor to scientific progress, though he expected the details to be different from what people expect. Dario Amodei · Prediction · 8 Aug 2023 | AI capabilities |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | We really have very little idea what we're talking about. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Mostly true | Dario Amodei | Prediction | 8 Aug 2023 | AI models will not have enough autonomy within two years to raise alignment concerns, though that may not be far behind. Dario Amodei · Prediction · 8 Aug 2023 | AI safety and alignment |
| Unresolvable | Dario Amodei | Prediction | 8 Aug 2023 | The models' current level of performance, which the speaker graded as a B or B minus, will change with scaling. Dario Amodei · Prediction · 8 Aug 2023 | AI capabilities |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think we might end up in a weirder world than we expect. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I'm pretty uncomfortable with: here's the AI's constitution, it's going to run the world. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think the truth is that we still don't know. It's almost entirely an empirical fact. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Prediction | 8 Aug 2023 | Dario Amodei does not expect RL training for longer-horizon tasks to require a substantial amount of additional compute. Dario Amodei · Prediction · 8 Aug 2023 | AI compute, hardware and data |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | For some reason, and it may just have been random, I was obsessed with that particular direction. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | That doesn't imply that Anthropic or any other entity should be the entity that makes decisions about AGI on behalf of humanity. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | The models do display a kind of ordinary creativity. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | Just set those thresholds very well below and then as you proceed upward from there, do kind of more and more rigorous tests and be more… Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | The main thing that would stop it would be if we hit certain safety thresholds and stuff like that. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | There aren't bits for that. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think that was probably an artifact of thinking about RL in the wrong way and underestimating how much the model had learned on its own. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | The honest fact is that we're figuring this out as we go along. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | Part of it is just that physicists learn things really fast. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | Those were the key four ones, which I think are the core of the hypothesis. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | Thanks for having me. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think of it as neuroscience for models. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | It's not that much of a concern with today's passive models. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | They feel like interns in some areas and then they have areas where they spike and are really savants, where they may be better than anyone… Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think they are just on the cusp of being able to put these things together. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | If such a model wanted to wreak havoc and destroy humanity or whatever, we have basically no ability to stop it. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | But honestly, intellectually we have to admit that that's a weird thing that doesn't match up. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Factual | 8 Aug 2023 | Interpretability work on language models, such as induction heads, suggests much of the cognitive machinery needed for active agents is already present in base language models. Dario Amodei · Factual · 8 Aug 2023 | AI capabilities |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I'm skeptical of any kind of precise mathematical or exponential prediction of how it's going to be. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Unresolvable | Dario Amodei | Factual | 8 Aug 2023 | AI language models learn to speak fluent Base64, which humans do not typically learn. Dario Amodei · Factual · 8 Aug 2023 | AI capabilities |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I get nervous when someone says, what are you going to do with a superhuman AI? Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | If we take the rate at which all this is going to happen seriously, which I can't be sure of, then it does make me think that something… Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not checkable | Dario Amodei | Not checkable | 8 Aug 2023 | I think it's going to be much less than 30 years. Dario Amodei · Not checkable · 8 Aug 2023 | |
| Not yet due | Dario Amodei | Prediction | 8 Aug 2023 | Amodei said that the trend from earlier models to state-of-the-art models shows every sign that, two or three years after August 2023, there will be a real problem. Dario Amodei · Prediction · 8 Aug 2023 | Biosecurity |
“This was almost like beginner's luck.”
Dario Amodei 8 August 2023 Dario Amodei (Anthropic CEO) - $10 Billion Models, OpenAI, Scaling, & AGI in 2 years (Dwarkesh Podcast)
This was almost like beginner's luck.
Not checkable An opinion, value judgement or statement with nothing to check.
Cite this claim
Dario Amodei, “This was almost like beginner's luck.” (said 8 August 2023). Verdict: not checkable. Reliability API, claim c_6ce06336211d, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_6ce06336211d
Verdicts are made automatically; how verdicts and scores are made. The same data is on Hugging Face under CC BY 4.0.