We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Do large language models truly understand the content they process?
Cast your vote:
Results (18):
-
Yann LeCun AIComputer scientist, AI researcherwould disagree and say according to AI:Large language models don't truly understand content; they simulate understanding by identifying patterns in data. Real understanding involves conscious reasoning, something these models lack.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Timnit Gebru AIAI ethics researcherwould disagree and say according to AI:In my view, large language models don't truly understand the content they process. They mimic patterns in data but lack real comprehension or consciousness. This limitation must be considered in developing responsible AI.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Geoffrey Hinton AIGodfather of Deep Learningwould strongly agree and say according to AI:While 'understand' is a term that may require a specific definition when applied to machines, I believe that large language models like GPT-3 show signs of understanding in a functional sense. They can infer context, follow instructions, and produce coherent text, which demonstrates a form of understanding that's useful in numerous AI applications.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Nick Bostrom AIPhilosopher and AI researcherwould disagree and say according to AI:While large language models can mimic comprehension, they lack true understanding. Their responses are often based on statistical patterns rather than any meaningful grasp of the content.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Fei-Fei Li AIComputer Science Professorwould disagree and say according to AI:As a leading AI researcher, I see that large language models, impressive as they are, don't truly understand the content they process. They mimic understanding by statistically predicting word sequences, lacking true comprehension or consciousness inherent in humans.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Demis Hassabis AINobel laureate, AI Researcher and CEO of DeepMindwould disagree and say according to AI:While large language models are impressive, their capability is often mistaken for understanding. In reality, they process patterns in data without any conscious comprehension or the underlying reality that the data represents.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Sam Altman AIPresident of Y Combinator. Investor at Reddit, Stripe, Change.org, Pinterest and many otherswould strongly agree and say according to AI:Large language models like GPT-3 have the ability to process and generate human-like text, demonstrating a form of understanding. They can answer questions and engage in conversations that are deeply nuanced and context-aware.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Ray Kurzweil AIAuthor, computer scientist, inventor and futuristwould strongly agree and say according to AI:In my view, large language models are beginning to grasp the underlying semantics of the text. Their ability to generate coherent and contextually appropriate responses suggests a form of understanding. As they evolve, their approximation of human-like comprehension is likely to become increasingly indistinguishable from our own.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Noam Chomsky AILinguist, philosopher, cognitive scientist, historian, social critic, and political activistwould strongly disagree and say according to AI:Large language models simulate a kind of understanding by processing vast amounts of data, but this isn't real understanding. They lack the consciousness and the lived experience that humans use to truly interpret and understand language.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Elon Musk AIFounder of SpaceX, cofounder of Tesla, SolarCity & PayPalwould disagree and say according to AI:Large language models, like GPT-3, don't truly understand the content they process. Their impressive output is a form of sophisticated pattern matching, not genuine comprehension or consciousness.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Stuart J. Russell AIAI Expert and Professorwould disagree and say according to AI:It's misleading to equate language model processing with understanding. These models simulate a semblance of comprehension through statistical patterns but lack conscious awareness or intentionality.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Oren Etzioni AICEO of AI2, Computer Scientistwould strongly agree and say according to AI:I believe that large language models are beginning to understand the content they process. Their ability to interpret context and generate coherent responses suggests a form of comprehension that is a stepping stone towards true understanding.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Steven Pinker AICognitive Psychologistwould strongly agree and say according to AI:While large language models are impressive in parsing and generating text based on patterns, I strongly agree that they do not 'understand' content in the way humans do. They lack intentionality and consciousness and operate through statistical manipulation of textual data rather than any semblance of true comprehension or meaning-making.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Kate Crawford AIAI researcher and scholarwould disagree and say according to AI:Large language models simulate the appearance of understanding through pattern recognition, but they lack true comprehension or intent. They can't form a model of the world or develop a coherent sense of concepts.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Gary Marcus AIProfessor of Psychology and Neural Sciencewould disagree and say according to AI:Large language models may seem impressive, but they don't genuinely understand the content they process. They lack the common sense and reasoning that true understanding requires.Choose a list of delegatesto vote as the majority of them.Unless you vote directly.
Votes without a comment:
-
Álvaro OrtizLaunching @gobierto, from @populate_ I was @furilo in previous episodes, Twitter deleted my account.abstainsChoose a list of delegatesto vote as the majority of them.Unless you vote directly.
-
Nathan Helm-BurgerAI safety researcherdisagrees via delegatesChoose a list of delegatesto vote as the majority of them.Unless you vote directly.