Claims

2,653 claims by 33 public figures and organisations, checked against the public record. Each has its quote, source, verdict and evidence. Read them without a key; browse by person.

28,548 claims, sorted by newest said. Choose a claim to read it in full.
VerdictPersonKindSaidClaimTopic
Unresolvable Yann LeCun Factual 14 Aug 2020 The Wall Street Journal published a piece on self-supervised learning involving Kyunghyun Cho and Yann LeCun. Yann LeCun · Factual · 14 Aug 2020 Their own work or organisation
Not researched Yann LeCun Factual 14 Aug 2020 Facebook AI Research (FAIR) has produced trainable Minecraft bots. Yann LeCun · Factual · 14 Aug 2020 AI capabilities
Mostly true Rodney Brooks Factual 14 Aug 2020 Rodney Brooks has said for years that adoption of autonomous vehicles will require changes to road infrastructure. Rodney Brooks · Factual · 14 Aug 2020 Technology
Not checkable Yann LeCun Not checkable 13 Aug 2020 Another advance for normalizing flows in high-energy physics simulations. Yann LeCun · Not checkable · 13 Aug 2020
Unresolvable Yann LeCun Factual 9 Aug 2020 The double descent test-error peak does not appear with multilayer neural networks because they are implicitly regularized. Yann LeCun · Factual · 9 Aug 2020 Science and research methods
Not checkable Yann LeCun Not checkable 9 Aug 2020 Very nice, intuitive and visual explanation for the "double descent" phenomenon that produces a peak in test error when the model capacity… Yann LeCun · Not checkable · 9 Aug 2020
Not checkable Yann LeCun Not checkable 7 Aug 2020 I think "jiga" is somewhat closer to the Greek pronounciation of γίγας Yann LeCun · Not checkable · 7 Aug 2020
Not checkable Rodney Brooks Not checkable 4 Aug 2020 It is good to be aware of the literature. Rodney Brooks · Not checkable · 4 Aug 2020
Not researched Rodney Brooks Citation 4 Aug 2020 Shannon and McCarthy's introduction (p. vi) to Automata Studies (Princeton University Press, 1956) describes a 'neural network' doing what GPT-3 does. Rodney Brooks · Citation · 4 Aug 2020 AI capabilities
Not checkable Rodney Brooks Not checkable 3 Aug 2020 I view them as possibilities rather than probabilities. Rodney Brooks · Not checkable · 3 Aug 2020
Unresolvable Rodney Brooks Factual 3 Aug 2020 On 2020-08-03 Rodney Brooks published a short blog post on how quickly accepted wisdom in science changes, with suggestions of possible radical scientific reorientations. Rodney Brooks · Factual · 3 Aug 2020 Their own work or organisation
Not checkable Marc Andreessen Not checkable 2 Aug 2020 quite possibly the best movie ever made Marc Andreessen · Not checkable · 2 Aug 2020
Not checkable Marc Andreessen Not checkable 2 Aug 2020 RIP Wilford Brimley, star of quite possibly the best movie ever made. Marc Andreessen · Not checkable · 2 Aug 2020
Not checkable Sam Altman Not checkable 1 Aug 2020 What matters is eventually hitting a home run, and the way to do that is training hard and a lot of at-bats. Sam Altman · Not checkable · 1 Aug 2020
Not checkable Sam Altman Not checkable 1 Aug 2020 Giving capital to promising people “too early” in their career is a great idea with much further to go, and the power law provides an… Sam Altman · Not checkable · 1 Aug 2020
Not checkable Sam Altman Not checkable 1 Aug 2020 You only have to be right once, and it's ok to be wrong a lot of times. Sam Altman · Not checkable · 1 Aug 2020
Not checkable Sam Altman Not checkable 1 Aug 2020 Be bold. Move fast. Work hard. Ignore haters. Keep swinging. Sam Altman · Not checkable · 1 Aug 2020
Not checkable Rodney Brooks Not checkable 31 Jul 2020 And the message is pretty damn powerful too. Rodney Brooks · Not checkable · 31 Jul 2020
Not checkable Rodney Brooks Not checkable 31 Jul 2020 This Nike ad is a technological tour de force which shows the impact of research in computer vision over the last fifty years. Rodney Brooks · Not checkable · 31 Jul 2020
Unresolvable Sam Altman Factual 29 Jul 2020 OpenAI's Scholars program provides stipends and mentorship to people from underrepresented groups to study deep learning full-time, now with 10 participants and a 6-month duration. Sam Altman · Factual · 29 Jul 2020 Field building and careers
Mostly true Sam Altman Factual 29 Jul 2020 OpenAI has launched applications for its Fall 2020 Scholars program. Sam Altman · Factual · 29 Jul 2020 Field building and careers
Not checkable Yann LeCun Not checkable 28 Jul 2020 Oh, and meters and kg not feet and pounds. Yann LeCun · Not checkable · 28 Jul 2020
Not checkable Yann LeCun Not checkable 28 Jul 2020 Yeah, we also say retropropagation not backpropagation, "convolution à trous" not "dilated convolution", "Dirac distribution" not "delta… Yann LeCun · Not checkable · 28 Jul 2020
Not checkable Rodney Brooks Not checkable 27 Jul 2020 Such a builder. Rodney Brooks · Not checkable · 27 Jul 2020
Not checkable Rodney Brooks Not checkable 25 Jul 2020 Good 20 minute piece on state of Tesla sef driving technology, out today from #CNBC. Rodney Brooks · Not checkable · 25 Jul 2020
Not checkable Yann LeCun Not checkable 25 Jul 2020 I have an overly complicated joke about convolutions but it's simpler in the Fourier domain. Yann LeCun · Not checkable · 25 Jul 2020
Not checkable Sam Altman Not checkable 24 Jul 2020 One of @paulg 's best (and most beautifully written!) essays: Sam Altman · Not checkable · 24 Jul 2020
Resolved true Sam Altman Factual 22 Jul 2020 Altman said that "we" share concerns about bias and safety in language models, and that this is a big part of why "we" are starting off with a beta and have safety review before apps can go live. Sam Altman · Factual · 22 Jul 2020 AI policy and governance
Not checkable Sam Altman Not checkable 22 Jul 2020 It's great to get feedback from someone with so much experience deploying AI at scale. Sam Altman · Not checkable · 22 Jul 2020
Not checkable Sam Altman Not checkable 20 Jul 2020 For non-programmers, it's like experiencing the magic of programming for the first time. Sam Altman · Not checkable · 20 Jul 2020
Not checkable Sam Altman Not checkable 20 Jul 2020 Theory: a big part of what resonates about the GPT-3 API is that you program it with English, so any English speaker can use it, and it… Sam Altman · Not checkable · 20 Jul 2020
Not checkable Sam Altman Not checkable 19 Jul 2020 It’s impressive (thanks for the nice compliments!) but it still has serious weaknesses and sometimes makes very silly mistakes. Sam Altman · Not checkable · 19 Jul 2020
Not checkable Sam Altman Not checkable 19 Jul 2020 We have a lot still to figure out. Sam Altman · Not checkable · 19 Jul 2020
Not checkable Sam Altman Not checkable 19 Jul 2020 The GPT-3 hype is way too much. Sam Altman · Not checkable · 19 Jul 2020
Not checkable Sam Altman Not checkable 19 Jul 2020 AI is going to change the world, but GPT-3 is just a very early glimpse. Sam Altman · Not checkable · 19 Jul 2020
Not checkable Rodney Brooks Not checkable 18 Jul 2020 Rename the bridge. Rodney Brooks · Not checkable · 18 Jul 2020
Not checkable Sam Altman Not checkable 18 Jul 2020 Wow just got access to GPT-3 and *my mind is blown*. Sam Altman · Not checkable · 18 Jul 2020
Not checkable Rodney Brooks Not checkable 17 Jul 2020 But then, I've never claimed to understand AI. Rodney Brooks · Not checkable · 17 Jul 2020
Not checkable Rodney Brooks Not checkable 17 Jul 2020 That's funny, I can't help but feel like GPT-3 is a way smaller deal than the hypnotists think. Rodney Brooks · Not checkable · 17 Jul 2020
Not checkable Sam Altman Not checkable 17 Jul 2020 Lots of nice feedback on the OpenAI API but the best yet is from a friend's 13 year old kid: Sam Altman · Not checkable · 17 Jul 2020
Not checkable Marc Andreessen Not checkable 16 Jul 2020 Now feels like the perfect time to fully express my views on a bunch of highly controversial topics! Marc Andreessen · Not checkable · 16 Jul 2020
Not checkable Rodney Brooks Not checkable 15 Jul 2020 These are real tweets. Rodney Brooks · Not checkable · 15 Jul 2020
Not checkable Sam Altman Not checkable 13 Jul 2020 A lot of great ideas look easy and obvious in retrospect, but it's still somehow enormously difficult to figure out the first time someone… Sam Altman · Not checkable · 13 Jul 2020
Not checkable Sam Altman Not checkable 13 Jul 2020 It's an easy way to trivialize important contributions. Sam Altman · Not checkable · 13 Jul 2020
Resolved true Rodney Brooks Prediction 9 Jul 2020 There will be no level 5 autonomous cars on roads in 2020. Rodney Brooks · Prediction · 9 Jul 2020 Technology
Not checkable Rodney Brooks Not checkable 9 Jul 2020 What Elon says here is false. Rodney Brooks · Not checkable · 9 Jul 2020
Not checkable Sam Altman Not checkable 8 Jul 2020 Make mistakes of ambition and not mistakes of sloth. Sam Altman · Not checkable · 8 Jul 2020
Resolved true Sam Altman Factual 8 Jul 2020 The deadline to apply for Apollo funding is the Saturday following 2020-07-08 (2020-07-11). Sam Altman · Factual · 8 Jul 2020 Field building and careers
Not checkable Sam Altman Not checkable 6 Jul 2020 There are two ways to make the future better, which everyone wants. Sam Altman · Not checkable · 6 Jul 2020
Not checkable Rodney Brooks Not checkable 6 Jul 2020 Strongly recommend Brian Cantwell Smith's optimistic "The Promise of Artificial Intelligence: Reckoning and Judgment", MIT Press 2019, for… Rodney Brooks · Not checkable · 6 Jul 2020
Unresolvable Rodney Brooks Factual 6 Jul 2020 As of July 2020, work on machine 'judgment' has not begun. Rodney Brooks · Factual · 6 Jul 2020 AI capabilities
Not checkable Rodney Brooks Not checkable 6 Jul 2020 Much easier (though still deeply intellectual) read than his previous writings. Rodney Brooks · Not checkable · 6 Jul 2020
Unresolvable Rodney Brooks Prediction 6 Jul 2020 AI systems doing reckoning will surpass human capability. Rodney Brooks · Prediction · 6 Jul 2020 AI capabilities
Not researched Rodney Brooks Citation 6 Jul 2020 Smith distinguishes 'reckoning' from 'judgment'. Rodney Brooks · Citation · 6 Jul 2020 AI capabilities
Not checkable Sam Altman Not checkable 6 Jul 2020 They lead to surprisingly different philosophies and actions. Sam Altman · Not checkable · 6 Jul 2020
Unresolvable Rodney Brooks Prediction 6 Jul 2020 Reckoning-type AI systems will have enormous scientific and economic impact. Rodney Brooks · Prediction · 6 Jul 2020 AI, the economy and work
Not checkable Sam Altman Not checkable 5 Jul 2020 The internet is disastrous for large-scale cohesion and amazing for small-scale community. Sam Altman · Not checkable · 5 Jul 2020
Not checkable Sam Altman Not checkable 5 Jul 2020 A super narrow rally is inconsistent with Americanness. Sam Altman · Not checkable · 5 Jul 2020
Not checkable Sam Altman Not checkable 5 Jul 2020 The most certain way for the US to fail is if we continue to fail to figure out how to share the economic benefits of progress broadly. Sam Altman · Not checkable · 5 Jul 2020
Not checkable Sam Altman Not checkable 5 Jul 2020 Amazing, and the only CEO of a megacompany that would do this Sam Altman · Not checkable · 5 Jul 2020
Not checkable Sam Altman Not checkable 5 Jul 2020 But the vision of shared success is what holds us together. Sam Altman · Not checkable · 5 Jul 2020
Resolved true Sam Altman Factual 5 Jul 2020 The US spends the most in the world on healthcare, by far. Sam Altman · Factual · 5 Jul 2020 Economics and markets
Not checkable Sam Altman Not checkable 5 Jul 2020 I always thought that was a good trade but now I don’t. Sam Altman · Not checkable · 5 Jul 2020
Not checkable Sam Altman Not checkable 5 Jul 2020 The US admits COVID-19 defeat while other developed countries go back to normal. Sam Altman · Not checkable · 5 Jul 2020
Not checkable Sam Altman Not checkable 5 Jul 2020 There are other things we have to do to maintain our position--import the world's best talent, invest super heavily in science and… Sam Altman · Not checkable · 5 Jul 2020
Not checkable Sam Altman Not checkable 4 Jul 2020 although deeply flawed, the usa is the best country in the world don’t @ me Sam Altman · Not checkable · 4 Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 Pandemics do therefore represent a source of existential risk. Nassim Taleb · Not checkable · Jul 2020
Unresolvable Nassim Taleb Factual Jul 2020 The United States and many other countries have spent trillions of dollars on sophisticated weaponry in past decades to counter uncertain threats. Nassim Taleb · Factual · Jul 2020 Philosophy, ethics and society
Not checkable Nassim Taleb Not checkable Jul 2020 probabilities are thin-tailed (since they are bounded by 0 and 1) but the corresponding payoff is not, so small errors in probability… Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 Understanding them actually reduces costs of reaction because they tell us what to target–and when to do so. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 In fact, distributional forecasts are more than hard–and often uninformative. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 It is not rational to worry about pandemic costs (extremely fat-tailed exposure), while not also questioning other sizable insurance-style… Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 Science is not about making single points predictions but about understanding properties Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 never underestimate consumers’ (nonlinear) behavior. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 It would be a good idea to question these expenditures first, before doubting the spending to stave off certain pandemics. Nassim Taleb · Not checkable · Jul 2020
Not researched Nassim Taleb Citation Jul 2020 Cirillo and Taleb, using extreme value theory, showed that pandemic deaths are fat-tailed. Nassim Taleb · Citation · Jul 2020 Biosecurity
Not checkable Nassim Taleb Not checkable Jul 2020 Forecasting single variables in fat-tailed domains is in violation of both common sense and probability theory. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 In some situations of fast-acting LLN, as (sometimes) in physics, properties can be revealed by single predictive experiments. Nassim Taleb · Not checkable · Jul 2020
Unresolvable Nassim Taleb Factual Jul 2020 In practically every location affected by COVID-19 there has been local or governmental action to mitigate the pandemic, as of July 2020. Nassim Taleb · Factual · Jul 2020 Biosecurity
Not checkable Nassim Taleb Not checkable Jul 2020 It is therefore incorrect to claim that it is the authorities’ response to the pandemic that caused unemployment in the transportation or… Nassim Taleb · Not checkable · Jul 2020
Not researched Nassim Taleb Citation Jul 2020 The "dismal" theorem (ref. 39) shows that trade-off analysis is an error under existential risk. Nassim Taleb · Citation · Jul 2020 Other global risks
Not checkable Nassim Taleb Not checkable Jul 2020 It is therefore preferable to look at other quantities, like for example the tail exponent. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 This is also wrong, because both forecasters (who missed) and their critics (complaining) were wrong. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 For matters of survival, particularly when systemic, and in the presence of multiplicative processes (like a pandemic), we require Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 Actually, the standard technique there used is the exact opposite of selection bias Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 The tail α is a direct function of the variance: the higher the variance of r, the thicker the tail of X. Nassim Taleb · Not checkable · Jul 2020
Unresolvable Nassim Taleb Factual Jul 2020 The dataset in reference 6 covers pandemics with fatalities above 1,000 and likely contains most, if not all, such pandemics. Nassim Taleb · Factual · Jul 2020 Biosecurity
Not checkable Nassim Taleb Not checkable Jul 2020 It is obvious that if forecasts lead to adjustments, and responses that affect the studied phenomenon, then one can no longer judge these… Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 the tails themselves (hence in the extremes), which can show remarkably stable properties. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 It is more cogent to blame overoptimization than reaction to disease. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 Mitigation has another effect: to delay and temporize, while we can understand the properties of the disease. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 errors in r could be small (but surely not zero), but their impact will be explosive on X, because of exponentiation. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 Errors in growth rates of a disease increase the fatness of tails in the distribution of fatalities. Nassim Taleb · Not checkable · Jul 2020
Unresolvable Nassim Taleb Prediction Jul 2020 The scale of the pandemic might change, but its tail properties will remain invariant. Nassim Taleb · Prediction · Jul 2020 Biosecurity
Not checkable Nassim Taleb Not checkable Jul 2020 if you must panic, it pays to panic early. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 In front of potentially fat-tailed random variables, more evidence is not necessarily needed. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 you cannot use naive B-school costs-benefit analyses for Russian roulette, because of the presence of an absorption barrier [32]. Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 One should not treat the economy and the disease as separate independent items, particularly by viewing a naive trade-off between economic… Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 a forecast can become a target that is gamed by participants Nassim Taleb · Not checkable · Jul 2020
Not checkable Nassim Taleb Not checkable Jul 2020 There are indeed severe nonlinearities at play. Nassim Taleb · Not checkable · Jul 2020
Back to the list

Sam Altman

Open its page

Not checkable

“You only have to be right once, and it's ok to be wrong a lot of times.”

Sam Altman 1 August 2020 Post by @sama on X, 2020-08-01 (X)

You only have to be right once, and it's ok to be wrong a lot of times.

Not checkable An opinion, value judgement or statement with nothing to check.

Cite this claim

Sam Altman, “You only have to be right once, and it's ok to be wrong a lot of times.” (said 1 August 2020). Verdict: not checkable. Reliability API, claim c_c76a04052b23, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_c76a04052b23

Verdicts are made automatically; how verdicts and scores are made. The same data is on Hugging Face under CC BY 4.0.