Video wird geladen...

Video konnte nicht geladen werden

Zur Startseite

Joscha Bach says current AI doesn't need to be conscious to execute tasks, but it must simulate reality to understand it LLMs don't just predict words To solve complex problems, they build deep internal models of spatial reasoning and the human theory of mind "that is why AI can...

slow_developer's profile picture

Haider.

64,586 subscribers

17,743 Aufrufe • vor 6 Monaten •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Kommentare

Keine Kommentare verfügbar

Kommentare vom Original-Post werden hier angezeigt

Ähnliche Videos

Sam Altman says the perfect AI is “a very tiny model with superhuman reasoning, 1 trillion tokens of context, and access to every tool you can imagine.” It doesn't need to contain the knowledge - just the ability to think, search, simulate, and solve anything.
0:53

Sam Altman says the perfect AI is “a very tiny model with superhuman reasoning, 1 trillion tokens of context, and access to every tool you can imagine.” It doesn't need to contain the knowledge - just the ability to think, search, simulate, and solve anything.

vitrupo

775,853 Aufrufe • vor 1 Jahr

To build safer AI, we need to understand how models "think". 🧠 Enter Gemma Scope 2, a new set of tools to interpret Gemma 3: our family of lightweight open models. It can help researchers trace internal reasoning, debug complex behaviors and identify risks →
0:56

To build safer AI, we need to understand how models "think". 🧠 Enter Gemma Scope 2, a new set of tools to interpret Gemma 3: our family of lightweight open models. It can help researchers trace internal reasoning, debug complex behaviors and identify risks →

Google DeepMind

219,486 Aufrufe • vor 8 Monaten

Yesterday at #GoogleIO, we introduced Gemini 2.5 Pro Deep Think 🧠, pushing the frontiers of AI reasoning. This enhanced reasoning mode is built to tackle drastically complex problems – like USAMO problems that stumped previous models. Super proud of the GDM team for this one!
0:30

Yesterday at #GoogleIO, we introduced Gemini 2.5 Pro Deep Think 🧠, pushing the frontiers of AI reasoning. This enhanced reasoning mode is built to tackle drastically complex problems – like USAMO problems that stumped previous models. Super proud of the GDM team for this one!

Oriol Vinyals

86,860 Aufrufe • vor 1 Jahr

Yann LeCun says LLMs are strongest in domains where language itself is the substrate of reasoning, like math and code They can solve problems, prove theorems, and write programs — but they are not creative mathematicians, software architects, or computer scientists "their role is to help humans build"
1:28

Yann LeCun says LLMs are strongest in domains where language itself is the substrate of reasoning, like math and code They can solve problems, prove theorems, and write programs — but they are not creative mathematicians, software architects, or computer scientists "their role is to help humans build"

Haider.

347,834 Aufrufe • vor 4 Monaten

In our latest AI Explainer episode, Sr. Product Marketing Manager for AI, Meena Ganesh, and Box CTO, Ben Kus, tackle beliefs holding enterprises back today. They dive into: • Model Hosting: Why "hosting it yourself" isn't the only path to enterprise-grade security. • The Reality of AI Tricking: A look at prompt injection and the "action guardrails" needed to protect your agents. • The "Clean Data" Fallacy: Why you don't need perfect data to start. • Determinism vs. Utility: Understanding why the "unpredictability" of LLMs is a feature for reasoning tasks. • Training vs. RAG: Why the models of today rarely need custom training sets.
2:05

In our latest AI Explainer episode, Sr. Product Marketing Manager for AI, Meena Ganesh, and Box CTO, Ben Kus, tackle beliefs holding enterprises back today. They dive into: • Model Hosting: Why "hosting it yourself" isn't the only path to enterprise-grade security. • The Reality of AI Tricking: A look at prompt injection and the "action guardrails" needed to protect your agents. • The "Clean Data" Fallacy: Why you don't need perfect data to start. • Determinism vs. Utility: Understanding why the "unpredictability" of LLMs is a feature for reasoning tasks. • Training vs. RAG: Why the models of today rarely need custom training sets.

Box

460,230 Aufrufe • vor 6 Monaten

📁 Terence Tao, mathematician and Fields Medalist, says AI may look like it reasons, but it does not understand. Language models imitate the steps of human reasoning without solid foundations. Sometimes they get it right, sometimes they collapse entirely. The output looks convincing, but it is fragile. It feels like understanding, yet it is probability pretending to think.
0:40

📁 Terence Tao, mathematician and Fields Medalist, says AI may look like it reasons, but it does not understand. Language models imitate the steps of human reasoning without solid foundations. Sometimes they get it right, sometimes they collapse entirely. The output looks convincing, but it is fragile. It feels like understanding, yet it is probability pretending to think.

Jon Hernandez

252,608 Aufrufe • vor 8 Monaten

Demis Hassabis, CEO of Google DeepMind, drops a quiet bombshell: The big question isn’t whether AI can solve problems. It’s whether AI can invent new science. Right now, it can’t. Not because of compute. Not because of data. But because it lacks something fundamental: A world model. Today’s LLMs can generate brilliant text, images, even code. But they don’t truly understand causality. They don’t know why A leads to B. They just predict patterns. Hassabis argues that real scientific discovery requires more: – Long-term planning – Stronger reasoning – And an internal model of how the world works Physics. Biology. Cause and effect. Only then can an AI run its own thought experiments. Only then do we get a true digital scientist.
1:47

Demis Hassabis, CEO of Google DeepMind, drops a quiet bombshell: The big question isn’t whether AI can solve problems. It’s whether AI can invent new science. Right now, it can’t. Not because of compute. Not because of data. But because it lacks something fundamental: A world model. Today’s LLMs can generate brilliant text, images, even code. But they don’t truly understand causality. They don’t know why A leads to B. They just predict patterns. Hassabis argues that real scientific discovery requires more: – Long-term planning – Stronger reasoning – And an internal model of how the world works Physics. Biology. Cause and effect. Only then can an AI run its own thought experiments. Only then do we get a true digital scientist.

VraserX e/acc

167,300 Aufrufe • vor 7 Monaten

Elon Musk explains why it's incredibly important for AI models to be grounded in reality "Physics is the law, and everything else is a recommendation. People break human-made laws, but no one breaks the laws of physics" AI must be grounded in reality to be truthful and accurate Examples: Cars: Must drive safely and correctly Optimus: Must perform real physical tasks A somewhat overlooked point: To be really intelligent, AI has to make predictions in line with reality (in other words, physics)
0:54

Elon Musk explains why it's incredibly important for AI models to be grounded in reality "Physics is the law, and everything else is a recommendation. People break human-made laws, but no one breaks the laws of physics" AI must be grounded in reality to be truthful and accurate Examples: Cars: Must drive safely and correctly Optimus: Must perform real physical tasks A somewhat overlooked point: To be really intelligent, AI has to make predictions in line with reality (in other words, physics)

X Freeze

26,288 Aufrufe • vor 5 Monaten

ELON MUSK: IT'S INCREDIBLY IMPORTANT FOR AI MODELS TO BE GROUNDED IN REALITY - "Physics is the law, and everything else is a recommendation" - People break human-made laws, but no one breaks the laws of physics - AI must be grounded in reality to ensure it is truthful and accurate - Examples of this necessity: Cars: Must drive safely and correctly. Optimus (Humanoid Robot): Must perform the physical tasks it is asked to do - Adhering to the laws of physics is helpful for ensuring the model is truthful - A somewhat overlooked point: To be really intelligent, AI has to make predictions that are in line with reality
1:08

ELON MUSK: IT'S INCREDIBLY IMPORTANT FOR AI MODELS TO BE GROUNDED IN REALITY - "Physics is the law, and everything else is a recommendation" - People break human-made laws, but no one breaks the laws of physics - AI must be grounded in reality to ensure it is truthful and accurate - Examples of this necessity: Cars: Must drive safely and correctly. Optimus (Humanoid Robot): Must perform the physical tasks it is asked to do - Adhering to the laws of physics is helpful for ensuring the model is truthful - A somewhat overlooked point: To be really intelligent, AI has to make predictions that are in line with reality

X Freeze

45,446 Aufrufe • vor 9 Monaten

Whatever AI sceptics say, LLMs really can reason. They're not just doing an imitation that looks like reasoning, it's the real deal. But even though they are able to reason, sometimes they won't! If you ask an LLM a question it can't answer, sometimes it will just try to imitate reasoning without doing it. The chain of thought looks basically indistinguishable from actual reasoning. But under the hood something very different is going on. Trenton Bricken talked with me about what work on circuits inside LLMs has revealed:
1:41

Whatever AI sceptics say, LLMs really can reason. They're not just doing an imitation that looks like reasoning, it's the real deal. But even though they are able to reason, sometimes they won't! If you ask an LLM a question it can't answer, sometimes it will just try to imitate reasoning without doing it. The chain of thought looks basically indistinguishable from actual reasoning. But under the hood something very different is going on. Trenton Bricken talked with me about what work on circuits inside LLMs has revealed:

Dwarkesh Patel

104,954 Aufrufe • vor 3 Monaten

GPT-5 on Sudoku-Bench 🧩 Since releasing Sudoku-Bench in May 2025, when no LLM could solve a classic 9x9 puzzle, we've been evaluating the latest generation of models. GPT-5 now leads our leaderboard with 33% puzzles solved--approximately 2x the previous leader--and is the first LLM we've tested to solve a 9x9 Sudoku variant. However, with 67% of the much harder puzzles remaining unsolved, Sudoku-Bench continues to present significant challenges for AI reasoning. Modern Sudoku variants require models to first understand novel rulesets through meta-reasoning, then maintain global consistency across long reasoning chains. Our experiments with GRPO fine-tuning on Qwen2.5-7b and "Thought Cloning" (training on expert human reasoning from Cracking the Cryptic) show that current approaches still struggle with the spatial reasoning and creative "break-in" points that human solvers use naturally. We believe new approaches are required to solve our benchmark. These results highlight persistent gaps between computational problem-solving and human-like reasoning, particularly in tasks requiring integrated mathematical logic, spatial awareness, and creative insight. Read more about our update here: 🔗 Blogpost →
0:28

GPT-5 on Sudoku-Bench 🧩 Since releasing Sudoku-Bench in May 2025, when no LLM could solve a classic 9x9 puzzle, we've been evaluating the latest generation of models. GPT-5 now leads our leaderboard with 33% puzzles solved--approximately 2x the previous leader--and is the first LLM we've tested to solve a 9x9 Sudoku variant. However, with 67% of the much harder puzzles remaining unsolved, Sudoku-Bench continues to present significant challenges for AI reasoning. Modern Sudoku variants require models to first understand novel rulesets through meta-reasoning, then maintain global consistency across long reasoning chains. Our experiments with GRPO fine-tuning on Qwen2.5-7b and "Thought Cloning" (training on expert human reasoning from Cracking the Cryptic) show that current approaches still struggle with the spatial reasoning and creative "break-in" points that human solvers use naturally. We believe new approaches are required to solve our benchmark. These results highlight persistent gaps between computational problem-solving and human-like reasoning, particularly in tasks requiring integrated mathematical logic, spatial awareness, and creative insight. Read more about our update here: 🔗 Blogpost →

Sakana AI

154,899 Aufrufe • vor 10 Monaten

“An AI that can act before it understands is not reliable infrastructure.” Yann LeCun, Meta’s Chief AI Scientist and one of the fathers of modern deep learning, says current LLMs are intrinsically unsafe. Not because they are evil, but because they hallucinate, lack common sense, and cannot reliably predict the consequences of agentic actions.
2:49

“An AI that can act before it understands is not reliable infrastructure.” Yann LeCun, Meta’s Chief AI Scientist and one of the fathers of modern deep learning, says current LLMs are intrinsically unsafe. Not because they are evil, but because they hallucinate, lack common sense, and cannot reliably predict the consequences of agentic actions.

Jon Hernandez

10,497 Aufrufe • vor 3 Monaten

Geoffrey Hinton says AI reasoning can be real thought because language itself is a form of thinking Words let humans and AI model almost anything, but human thought goes beyond words — through images, space, and physical movement "the smarter system is the one that can use all of them"
1:13

Geoffrey Hinton says AI reasoning can be real thought because language itself is a form of thinking Words let humans and AI model almost anything, but human thought goes beyond words — through images, space, and physical movement "the smarter system is the one that can use all of them"

Haider.

24,880 Aufrufe • vor 4 Monaten

Yann LeCun says robotics could become a huge market, but the industry still doesn't know how to make humanoid robots useful The missing piece is AI that can understand the physical world and handle tasks with generality and adaptability "LLMs are useful, but for different problems"
1:00

Yann LeCun says robotics could become a huge market, but the industry still doesn't know how to make humanoid robots useful The missing piece is AI that can understand the physical world and handle tasks with generality and adaptability "LLMs are useful, but for different problems"

Haider.

13,727 Aufrufe • vor 3 Monaten

Emad Mostaque says by next year, AI reasoning models like OpenAI's o1 and DeepSeek's R1 will run on smartphones and perform PhD-level tasks with 20 watts of electricity - equivalent to the human brain
1:26

Emad Mostaque says by next year, AI reasoning models like OpenAI's o1 and DeepSeek's R1 will run on smartphones and perform PhD-level tasks with 20 watts of electricity - equivalent to the human brain

Tsarathustra

157,136 Aufrufe • vor 1 Jahr

Without World Models, There Is No AGI. Google Just Proved It. If AGI ever happens, it will not come from bigger chatbots alone. From the very start of this interview, one thing is crystal clear: without world models, we will never reach AGI. And right now, Google is leading with its world simulator Genie 3. Here is the core of what Demis Hassabis explains in this conversation: • World models are the missing core of AGI Hassabis says his deepest long term focus has always been world models and simulations. Not just language. Not just prediction. Actual internal simulations of reality. • LLMs are impressive, but incomplete Language models understand more about the world than expected because human language encodes a lot of reality. Still, language is only a shadow of the real thing. • What text can never fully teach Reality includes things text struggles to express: •3D space and spatial dynamics •Physical causality and mechanics •Sensorimotor experience like movement, force, smell, or balance • Experience beats description To close the gap, AI must learn from interaction and experience, not just static text. That is how you build an internal world simulator. • Why Genie 3 matters With Google DeepMind pushing systems like Genie 3, AI starts to model reality itself, not just talk about it. • Robots and real world assistants depend on this True robotics, smart glasses, and universal assistants require AI that understands the physical world you live in, not just your screen. Bottom line: AGI will not emerge from better text prediction. It will emerge from systems that can simulate, predict, and understand reality itself. Right now, Google is clearly ahead on that path. Curious what you think. Are world models the real AGI unlock, or just another stepping stone?
1:26

Without World Models, There Is No AGI. Google Just Proved It. If AGI ever happens, it will not come from bigger chatbots alone. From the very start of this interview, one thing is crystal clear: without world models, we will never reach AGI. And right now, Google is leading with its world simulator Genie 3. Here is the core of what Demis Hassabis explains in this conversation: • World models are the missing core of AGI Hassabis says his deepest long term focus has always been world models and simulations. Not just language. Not just prediction. Actual internal simulations of reality. • LLMs are impressive, but incomplete Language models understand more about the world than expected because human language encodes a lot of reality. Still, language is only a shadow of the real thing. • What text can never fully teach Reality includes things text struggles to express: •3D space and spatial dynamics •Physical causality and mechanics •Sensorimotor experience like movement, force, smell, or balance • Experience beats description To close the gap, AI must learn from interaction and experience, not just static text. That is how you build an internal world simulator. • Why Genie 3 matters With Google DeepMind pushing systems like Genie 3, AI starts to model reality itself, not just talk about it. • Robots and real world assistants depend on this True robotics, smart glasses, and universal assistants require AI that understands the physical world you live in, not just your screen. Bottom line: AGI will not emerge from better text prediction. It will emerge from systems that can simulate, predict, and understand reality itself. Right now, Google is clearly ahead on that path. Curious what you think. Are world models the real AGI unlock, or just another stepping stone?

VraserX e/acc

23,784 Aufrufe • vor 8 Monaten