Tools

Explore

Videos Channels Figures

Atmrix

Tools

Explore

Videos Channels Figures

Atmrix

LF

Lex Fridman Cast

02/03/25

@ Lex Fridman

The word on the San Francisco Street is that Anthropic has a better model than o3, but they won't release it because chains of thought are scary, and they are legitimately scary.

Video

DeepSeek, China, OpenAI, NVIDIA, xAI, TSMC, Stargate, and AI Megaclusters | Lex Fridman Podcast #459

@ Lex Fridman

DeepSeek, China, OpenAI, NVIDIA, xAI, TSMC, Stargate, and AI Megaclusters | Lex Fridman Podcast #459

02/03/25

Related Takeaways

Y Combinator Cast

10/25/24

@ Y Combinator

OpenAI's approach to implementing chain of thought reasoning in o1 remains largely undisclosed, but it likely involves advanced prompt engineering techniques.

Y Combinator Cast

10/25/24

@ Y Combinator

Research shows that using Chain of Thought reasoning, o1 can solve inherently serial problems, indicating significant potential for this model series with sufficient computational resources.

Garry Tan

12/07/24

@ Y Combinator

In October, Anthropic released upgraded models including Claude 3.5 and introduced the innovative feature of computer use, positioning itself as a leader in the rapidly growing AI agent landscape.

Mike Krieger

05/09/25

@ Sequoia Capital

At Anthropic, we embrace a bottoms-up approach to creativity, allowing for more innovative product development close to the model's capabilities.

Y Combinator Cast

11/22/23

@ Y Combinator

The idea that an autonomous agent with thoughts and beliefs exists within a foundation model is misleading; I think that's not true.

Geoffrey Hinton

04/26/25

@ CBS Mornings

Anthropic is currently the most concerned with AI safety, especially after many safety researchers left OpenAI due to concerns about its direction.

Y Combinator Cast

10/25/24

@ Y Combinator

The models o1 preview and o1 mini represent a new class of models designed to reason through complex problems.

Y Combinator Cast

10/25/24

@ Y Combinator

The chain of thought strategy, introduced by Google Brain researchers in 2022, helps models like o1 to solve problems step by step.

Kyle Fish

04/24/25

@ Anthropic

I work on model welfare at Anthropic, exploring whether AI systems like Claude might have experiences we should consider and what actions we should take if they do.