Tools

Search

Explore

Videos Channels Figures

Atmrix

About

Tools

Search

Explore

Videos Channels Figures

Atmrix

About

GH

Geoffrey Hinton

01/18/25

@ Curt Jaimungal

Understanding in AI and humans involves associating features with words and how those features interact, rather than relying on a magical internal process.

Video

CJ

Why The "Godfather of AI" Now Fears His Own Creation | Geoffrey Hinton

@ Curt Jaimungal

Why The "Godfather of AI" Now Fears His Own Creation | Geoffrey Hinton

01/18/25

Related Takeaways

A

Anthropic Cast

03/27/25

@ Anthropic

We aim to understand how an AI model connects concepts in its mind to formulate answers to our questions.

JW

Josh Woodward

03/18/25

@ Sequoia Capital

The concept of long context in AI is crucial, as it allows for deeper understanding and communication, similar to personal relationships.

MK

Mike Krieger

05/31/25

@ Hard Fork

Anthropic's approach to AI involves training models to align with desired behaviors rather than relying solely on rigid rules, which can be insufficient in complex situations.

MS

Murray Shanahan

06/29/17

@ Y Combinator

Current AI technologies excel in specific tasks like image recognition and speech processing, but they still struggle with genuine understanding of language.

AB

Andrew Bosworth

04/24/25

@ a16z

The AI we have today can understand user intentions better, allowing for more natural interactions and outcomes based on vague references.

LF

Lex Fridman

05/29/25

@ Andrew Huberman

The interaction between humans and robots is crucial, as it will shape how we work together to accomplish tasks, and understanding this dynamic is essential for future advancements in AI.

YC

Y Combinator Cast

11/22/23

@ Y Combinator

Understanding AGI is challenging because we don't fully comprehend human intelligence or how it works.

LF

Lex Fridman Cast

02/03/25

@ Lex Fridman

The future of AI will depend on how well we can integrate human insights into AI systems, ensuring they are beneficial and aligned with human values.

DA

Dario Amodei

08/08/23

@ Dwarkesh Patel

Mechanistic interpretability aims to understand what's happening inside AI models at the level of individual circuits, but we still don't know enough to ensure alignment.