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
Mostly true Marc Andreessen Prediction 12 Jun 2023 Andreessen said he thought the rate of AI technological improvement from mid-2023 onward would be very rapid. Marc Andreessen · Prediction · 12 Jun 2023 AI timelines and forecasting
Mostly false Marc Andreessen Factual 12 Jun 2023 Outside the San Francisco Bay Area, AI development is overwhelmingly in Beijing and Shanghai. Marc Andreessen · Factual · 12 Jun 2023 AI compute, hardware and data
Not researched Marc Andreessen Factual 12 Jun 2023 As of June 2023, Germany is paying Russia for oil and gas, which Andreessen describes as an energy war with Russia. Marc Andreessen · Factual · 12 Jun 2023 Other global risks
Unresolvable Marc Andreessen Prediction 12 Jun 2023 Countries like Israel will have their own AI capabilities, but much of what they have will be derived from the United States. Marc Andreessen · Prediction · 12 Jun 2023 AI policy and governance
Not researched Marc Andreessen Factual 12 Jun 2023 The computer system he is describing knows about many more things, but that does not translate into a higher level of fluid intelligence. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Not researched Marc Andreessen Factual 12 Jun 2023 The neural network architecture used in GPT-4 is based on a paper originally written in 1943. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Not researched Marc Andreessen Factual 12 Jun 2023 The concentration of AI development in the Bay Area results from Stanford, Berkeley, and the AI labs of Google, Facebook and Microsoft having worked on it there for decades. Marc Andreessen · Factual · 12 Jun 2023 AI compute, hardware and data
Mixed Marc Andreessen Factual 12 Jun 2023 Hundreds of studies' conclusions on the benefits of intelligence all correlate with fluid intelligence, the ability to rapidly assimilate knowledge and use it to solve problems. Marc Andreessen · Factual · 12 Jun 2023 Philosophy, ethics and society
Resolved true Marc Andreessen Factual 12 Jun 2023 The AI being discussed is not, strictly speaking, human intelligence. Marc Andreessen · Factual · 12 Jun 2023 AI capabilities
Not checkable Marc Andreessen Not checkable 11 Jun 2023 BRITISH DYNAMISM Marc Andreessen · Not checkable · 11 Jun 2023
Not checkable Marc Andreessen Not checkable 11 Jun 2023 Ask Me Anything about AI, or any of the topics in Why AI Will Save The World, as responses to this tweet and I will answer as many as I can. Marc Andreessen · Not checkable · 11 Jun 2023
Not checkable Rodney Brooks Not checkable 10 Jun 2023 The short animation of AI killing off humanity by 3D printing robot vacs makes clicking on this worthwhile! Rodney Brooks · Not checkable · 10 Jun 2023
Not checkable Marc Andreessen Not checkable 10 Jun 2023 Enjoy this discussion with @martin_casado and me about how AI will save the world! Marc Andreessen · Not checkable · 10 Jun 2023
Not checkable Marc Andreessen Not checkable 9 Jun 2023 Read this my response to my essay Why AI Will Save The World from @JimPethokoukis! Marc Andreessen · Not checkable · 9 Jun 2023
Unresolvable Rodney Brooks Factual 8 Jun 2023 Brooks called Isambard Kingdom Brunel (1806-59) the world's greatest engineer, crediting him with the "longest/most beautiful bridges", the first iron-hulled screw ships, and England's tunnels and railways. Rodney Brooks · Factual · 8 Jun 2023 Other
Not checkable Marc Andreessen Not checkable 8 Jun 2023 Read this 2nd response to me by @titusfilm! Marc Andreessen · Not checkable · 8 Jun 2023
Unresolvable Balaji Srinivasan Factual 7 Jun 2023 As of June 2023, the balajis.com community chat is free to join. Balaji Srinivasan · Factual · 7 Jun 2023 Their own work or organisation
Not checkable Ørsted Not checkable 7 Jun 2023 Ørsted may reconfigure or exit awarded projects where Ørsted has as of this date not yet taken FID if it finds the value creation in the… Ørsted · Not checkable · 7 Jun 2023
Not checkable Marc Andreessen Not checkable 7 Jun 2023 Join us in 8 minutes! Marc Andreessen · Not checkable · 7 Jun 2023
Not researched Ørsted Factual 7 Jun 2023 Ørsted has significantly grown its onshore renewables business and become a leading player in bringing commercial scale to Power-to-X. Ørsted · Factual · 7 Jun 2023 Their own work or organisation
Not checkable Balaji Srinivasan Not checkable 7 Jun 2023 But to keep the community friendly, we reserve the right to (a) do content moderation and/or (b) require payment to comment. Balaji Srinivasan · Not checkable · 7 Jun 2023
Unresolvable Balaji Srinivasan Factual 7 Jun 2023 Balaji Srinivasan is consolidating his content into one free newsletter at balajis.com. Balaji Srinivasan · Factual · 7 Jun 2023 Their own work or organisation
Not checkable Ørsted Not checkable 7 Jun 2023 Ørsted has delivered strong progress since the CMD in 2021. Ørsted · Not checkable · 7 Jun 2023
Not yet due Ørsted Promise 7 Jun 2023 Ørsted has committed to reusing or recycling all wind turbine blades and solar panels from its global portfolio with immediate effect, and to commissioning only projects with a net-positive biodiversity impact from 2030. Ørsted · Promise · 7 Jun 2023 Their own work or organisation
Mostly true Rodney Brooks Factual 7 Jun 2023 Rodney Brooks has taken many fully driverless Cruise rides at night in San Francisco. Rodney Brooks · Factual · 7 Jun 2023 Technology
Not researched Ørsted Factual 7 Jun 2023 Ørsted has been awarded an additional 10.6 GW and has a renewable project pipeline of about 114 GW. Ørsted · Factual · 7 Jun 2023 Their own work or organisation
Unresolvable Ørsted Prediction 7 Jun 2023 Ørsted's long-standing supplier relationships will ensure it capacity at competitive prices. Ørsted · Prediction · 7 Jun 2023 Their own work or organisation
Not checkable Rodney Brooks Not checkable 7 Jun 2023 But still a long way from commercial success, and still pissing off many human drivers during the day Rodney Brooks · Not checkable · 7 Jun 2023
Mostly true Balaji Srinivasan Factual 7 Jun 2023 Balaji Srinivasan has created tweets, talks, podcasts, courses, articles and The Network State book. Balaji Srinivasan · Factual · 7 Jun 2023 Their own work or organisation
Not yet due Ørsted Prediction 7 Jun 2023 Ørsted is on track to outperform its previous EBITDA and ROCE targets for 2020-2027. Ørsted · Prediction · 7 Jun 2023 Their own work or organisation
Not checkable Balaji Srinivasan Not checkable 7 Jun 2023 even more specifically, how we build the network state. Balaji Srinivasan · Not checkable · 7 Jun 2023
Not yet due Ørsted Promise 7 Jun 2023 Ørsted reaffirms its target of about 50 GW of renewable capacity in 2030. Ørsted · Promise · 7 Jun 2023 Their own work or organisation
Resolved true Ørsted Factual 7 Jun 2023 Ørsted currently operates 15.5 GW of renewable energy assets, with 4.9 GW under construction. Ørsted · Factual · 7 Jun 2023 Their own work or organisation
Not researched Ørsted Factual 7 Jun 2023 Since the 2021 CMD, Ørsted has completed several large projects, including the world's largest offshore wind farm. Ørsted · Factual · 7 Jun 2023 Their own work or organisation
Not checkable Balaji Srinivasan Not checkable 7 Jun 2023 in addition to distributing NFTs (like this), I’d like to issue prizes (like this), invest in subscribers (like this), hold in-person… Balaji Srinivasan · Not checkable · 7 Jun 2023
Not yet due Ørsted Prediction 7 Jun 2023 Ørsted is on track to outperform its previous average ROCE target of 11-12% for 2020-2027, now expecting about 15% over 2020-2027. Ørsted · Prediction · 7 Jun 2023 Their own work or organisation
Not researched Ørsted Factual 7 Jun 2023 Ørsted is the global leader in offshore wind, with the largest portfolio of operating assets and the largest project pipeline across Europe, the US and APAC. Ørsted · Factual · 7 Jun 2023 Their own work or organisation
Not checkable Ørsted Not checkable 7 Jun 2023 we’ll only take FID on projects that we find will create sufficient value. Ørsted · Not checkable · 7 Jun 2023
Not yet due Ørsted Prediction 7 Jun 2023 Ørsted's gross investment for 2020-2027, previously expected at about DKK 350 billion, is trending higher at DKK 380 billion. Ørsted · Prediction · 7 Jun 2023 Their own work or organisation
Unresolvable Ørsted Factual 7 Jun 2023 Ørsted is recognised on the CDP Climate Change A List as a global leader on climate action. Ørsted · Factual · 7 Jun 2023 Their own work or organisation
Not checkable Balaji Srinivasan Not checkable 7 Jun 2023 That’s essays, talks, podcasts, prizes, and updates to The Network State book. Balaji Srinivasan · Not checkable · 7 Jun 2023
Not researched Ørsted Factual 7 Jun 2023 Ørsted is the world's first energy company to have issued blue bonds. Ørsted · Factual · 7 Jun 2023 Their own work or organisation
Resolved true Balaji Srinivasan Factual 7 Jun 2023 There is a community chat associated with the newsletter at balajis.com/chat. Balaji Srinivasan · Factual · 7 Jun 2023 Their own work or organisation
Not checkable Balaji Srinivasan Not checkable 7 Jun 2023 Well, it’s the long form version of everything I tweet about — but it’s specifically about the network state. Balaji Srinivasan · Not checkable · 7 Jun 2023
Not checkable Marc Andreessen Not checkable 7 Jun 2023 Read this response to me from @jamespoulos! Marc Andreessen · Not checkable · 7 Jun 2023
Resolved true Balaji Srinivasan Factual 7 Jun 2023 Balaji Srinivasan's newsletter audience has more than 100,000 subscribers. Balaji Srinivasan · Factual · 7 Jun 2023 Their own work or organisation
Not checkable Balaji Srinivasan Not checkable 7 Jun 2023 I’ve never thought of myself as a “content creator”, but I’ve certainly created content all over the place. Balaji Srinivasan · Not checkable · 7 Jun 2023
Not yet due Ørsted Prediction 7 Jun 2023 Ørsted now expects EBITDA from offshore and onshore assets in operation to grow at a 15% CAGR over 2020-2027, versus the previous target of about 12%. Ørsted · Prediction · 7 Jun 2023 Their own work or organisation
Own records only Ørsted Promise 7 Jun 2023 Ørsted will target an unlevered, fully loaded lifecycle IRR at a 150-300 basis point spread to WACC at the time of bid or FID, whichever comes first, while investing a majority of CAPEX in offshore. Ørsted · Promise · 7 Jun 2023 Their own work or organisation
Unresolvable Balaji Srinivasan Promise 7 Jun 2023 All content on the balajis.com newsletter will be free in perpetuity. Balaji Srinivasan · Promise · 7 Jun 2023 Their own work or organisation
Own records only Rodney Brooks Factual 7 Jun 2023 On Cruise rides in San Francisco, the car made left turns at speed with traffic that had close calls with parked cars. Rodney Brooks · Factual · 7 Jun 2023 Technology
Not yet due Ørsted Prediction 7 Jun 2023 Ørsted is on track to reach its science-based targets of 98% carbon emissions reductions (scopes 1 and 2) in 2025 and net-zero for the full value chain in 2040. Ørsted · Prediction · 7 Jun 2023 Their own work or organisation
Not checkable Balaji Srinivasan Not checkable 7 Jun 2023 It’s fun. Finally, I’d like to get creative with this by trying various experiments, some of which are iterations on things I’ve tried over… Balaji Srinivasan · Not checkable · 7 Jun 2023
Not checkable Balaji Srinivasan Not checkable 7 Jun 2023 we’ll be using this newsletter to help bootstrap startup societies and network states all around the world, like Culdesac, Prospera, Cabin… Balaji Srinivasan · Not checkable · 7 Jun 2023
Not checkable Marc Andreessen Not checkable 7 Jun 2023 Read this response to me from @titusfilm! Marc Andreessen · Not checkable · 7 Jun 2023
Resolved true Ørsted Factual 7 Jun 2023 Ørsted was the first energy company in the world to have its science-based net-zero emissions target validated by the Science Based Targets initiative. Ørsted · Factual · 7 Jun 2023 Their own work or organisation
Unresolvable Marc Andreessen Prediction 6 Jun 2023 Artificial intelligence will not destroy the world, and may instead save it. Marc Andreessen · Prediction · 6 Jun 2023 AI timelines and forecasting
Not checkable Marc Andreessen Not checkable 6 Jun 2023 Why AI Will Save The World -- my new megapost on why you should be excited, not scared, about AI. Enjoy! Marc Andreessen · Not checkable · 6 Jun 2023
Not checkable Marc Andreessen Not checkable 6 Jun 2023 Join us tonight -- 8PM Pacific -- for an Effective Accelerationism (E/Acc) discussion of why AI will save the world! Marc Andreessen · Not checkable · 6 Jun 2023
Not checkable Marc Andreessen Not checkable 6 Jun 2023 Enjoy! Marc Andreessen · Not checkable · 6 Jun 2023
Not researched Rodney Brooks Factual 5 Jun 2023 Pointing to an account by writer Cory Doctorow, Brooks said the simulated drone did not kill its simulated operator. Rodney Brooks · Factual · 5 Jun 2023 AI safety and alignment
Not checkable Rodney Brooks Not checkable 5 Jun 2023 But our evil sorcerer AI bros are mostly huxters. Rodney Brooks · Not checkable · 5 Jun 2023
Not checkable Rodney Brooks Not checkable 5 Jun 2023 Being an 'evil sorcerer' is more acceptable than being a 'huxter'. Rodney Brooks · Not checkable · 5 Jun 2023
Not checkable Rodney Brooks Not checkable 5 Jun 2023 Apparently he used AI to produce his paintings. Rodney Brooks · Not checkable · 5 Jun 2023
Not researched Yann LeCun Factual Jun 2023 The denoising autoencoder (ref. 51) is a type of contrastive energy-based model. Yann LeCun · Factual · Jun 2023 AI capabilities
Not researched Yann LeCun Factual Jun 2023 Supervised and reinforcement learning require an enormous number of labeled samples or trials. Yann LeCun · Factual · Jun 2023 AI capabilities
Not checkable Yann LeCun Not checkable Jun 2023 Human learning, in contrast, is often highly efficient. Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 However, which information is important may be scale-dependent. Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 The proposed solution is, therefore, to replace probabilistic models with energy-based models (EBMs) and combine them with latent variables… Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 In the case of the latent variable EBMs (Fig. 4(b)), such Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 Contrastive methods aim to push down the energy of training samples and to pull up the energy of suitably-generated contrastive samples… Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 Current automated systems have crucial limitations that need to be addressed before artificial intelligence can reach human-like levels and… Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 An important feature of latent variable models is that while they are composed of deterministic functions, they are non-deterministic as… Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 Therefore, the energy function is not the objective function to minimize within the learning! Yann LeCun · Not checkable · Jun 2023
Unresolvable Yann LeCun Factual Jun 2023 Contrastive methods for training energy-based models work for any architecture but scale badly with data dimension. Yann LeCun · Factual · Jun 2023 AI capabilities
Not researched Yann LeCun Factual Jun 2023 Energy-based models can represent multimodal dependencies in which multiple values of y are compatible with a given x. Yann LeCun · Factual · Jun 2023 AI capabilities
Mostly true Yann LeCun Factual Jun 2023 Masked autoencoders are less successful with images and other continuous domains than with discrete domains such as language. Yann LeCun · Factual · Jun 2023 AI capabilities
Unresolvable Yann LeCun Prediction Jun 2023 H-JEPA trained with regularized methods may be a starting point for predictive world models capable of hierarchical planning under uncertainty, a breakthrough for autonomous AI. Yann LeCun · Prediction · Jun 2023 AI timelines and forecasting
Not checkable Yann LeCun Not checkable Jun 2023 This proposal is important to the whole ML community Yann LeCun · Not checkable · Jun 2023
Not researched Yann LeCun Factual Jun 2023 Contrastive methods for training energy-based models tend to become very inefficient in high dimensions. Yann LeCun · Factual · Jun 2023 AI capabilities
Not checkable Yann LeCun Not checkable Jun 2023 SSL is the cake (provides millions of information bits per sample) Yann LeCun · Not checkable · Jun 2023
Not researched Yann LeCun Citation Jun 2023 Convolutional sparse autoencoders learn features corresponding to known image filters used for feature extraction and compressed representation (ref. 47). Yann LeCun · Citation · Jun 2023 AI capabilities
Not checkable Yann LeCun Not checkable Jun 2023 Firstly, it is challenging to understand uncertainty in an energy-based setting. Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 To prevent the energy collapse, i.e., flat energy landscape, we need to constrain our energy surface so that data points have low energies… Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 A joint embedding predictive architecture (JEPA), presented in Fig. 9, is an EBM that combines Yann LeCun · Not checkable · Jun 2023
Unresolvable Yann LeCun Prediction Jun 2023 Self-supervised learning is the most likely approach to learning world representations, since supervised and reinforcement learning require too many samples or trials. Yann LeCun · Prediction · Jun 2023 AI capabilities
Not checkable Yann LeCun Not checkable Jun 2023 Energy-based models (EBMs) capture dependencies between variables x and y by associating a scalar energy F (x, y) to each configuration of… Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 This pitfall is due to the folded structure of the data, which, however, rarely Yann LeCun · Not checkable · Jun 2023
Not researched Yann LeCun Factual Jun 2023 Babies learn almost entirely by observation. Yann LeCun · Factual · Jun 2023 Science and research methods
Not checkable Yann LeCun Not checkable Jun 2023 it can collapse because it can learn the identity function and have zero energy for all the inputs. Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 Such multi-level predictions may be achieved with a hierarchical joint embedding predictive architecture (H-JEPA). Yann LeCun · Not checkable · Jun 2023
Not researched Yann LeCun Factual Jun 2023 Hopfield networks, fully-connected recurrent networks, were popularized by John Hopfield in 1982. Yann LeCun · Factual · Jun 2023 AI capabilities
Not checkable Yann LeCun Not checkable Jun 2023 The tricky part is that the information capacity of the latent variable must be minimized. Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 Moreover, JEPAs naturally handle Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 The main disadvantage of maximum likelihood as a contrastive EBM is that it wants to make the difference between the energy on the data… Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 To avoid this drawback, we need to design an architecture that does not require transferring outputs between its components. Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 While it may be surprising that we give up on a probabilistic setting, note that making decisions can be viewed as choosing the option with… Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 In particular, energies are uncalibrated (i.e., measured in arbitrary units), so combining two separately trained EBMs is not… Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 Finally, the maximum likelihood can be interpreted as a special case of a contrastive method. Yann LeCun · Not checkable · Jun 2023
Not checkable Yann LeCun Not checkable Jun 2023 It means that the minimization of LNLL consists in minimizing the energy of training samples while pulling up the energy of Monte… Yann LeCun · Not checkable · Jun 2023
Back to the list

Yann LeCun

Open its page

Not checkable

“Current automated systems have crucial limitations that need to be addressed before artificial intelligence can reach human-like levels and bring new technological revolutions.”

Yann LeCun June 2023 Introduction to Latent Variable Energy-Based Models (Les Houches lecture notes) (arXiv)

Current automated systems have crucial limitations that need to be addressed before artificial intelligence can reach human-like levels and…

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

Cite this claim

Yann LeCun, “Current automated systems have crucial limitations that need to be addressed before artificial intelligence can reach human-like levels and…” (said June 2023). Verdict: not checkable. Reliability API, claim c_0e2bc49b128b, method 0.1.0+c876030c. https://epistacklabs.com/claims/c_0e2bc49b128b

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