Загрузка видео...

Не удалось загрузить видео

На главную

Severance as a show about interpretability research in AI (a clip from a talk; YouTube link just below):

ChrisGPotts's profile picture

Christopher Potts

16,222 subscribers

44,892 просмотров • 9 месяцев назад •via X (Twitter)

ОбразованиеНаука и технологии
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

Interpretability research has made only minor contributions to AI safety so far. What can we do to change that? (Clip from a longer talk; YouTube link in the thread):
1:50

Interpretability research has made only minor contributions to AI safety so far. What can we do to change that? (Clip from a longer talk; YouTube link in the thread):

Christopher Potts

29,211 просмотров • 9 месяцев назад

The Anthropic perspective on interpretability is prominent and significant, but not inevitable. My own take is quite different. (Clip from a talk I gave; YouTube link in the thread):
2:18

The Anthropic perspective on interpretability is prominent and significant, but not inevitable. My own take is quite different. (Clip from a talk I gave; YouTube link in the thread):

Christopher Potts

55,872 просмотров • 9 месяцев назад

I recently gave a talk at Yale Law School about writing AI constitutions. Video below and on YouTube, link to transcript and slides in thread.
1:28:45

I recently gave a talk at Yale Law School about writing AI constitutions. Video below and on YouTube, link to transcript and slides in thread.

Joe Carlsmith

40,652 просмотров • 4 месяцев назад

DAY 8 of making a feature film with AI in 30 days. Today I show a clip of my film and discuss directing through audio. As always, the prompts from today's video are in the description under the YouTube version linked below.
9:35

DAY 8 of making a feature film with AI in 30 days. Today I show a clip of my film and discuss directing through audio. As always, the prompts from today's video are in the description under the YouTube version linked below.

Jenny Krakovsky

24,474 просмотров • 2 месяцев назад

Her actions came from love, not desperation — a distinction many people don’t talk about. Full video still showing on YouTube, click the link below 👇
0:25

Her actions came from love, not desperation — a distinction many people don’t talk about. Full video still showing on YouTube, click the link below 👇

Instablog9ja

295,596 просмотров • 7 месяцев назад

DAY 7 of making a feature film with AI in 30 days. Today I show a clip of my film and discuss scene coverage, blocking, framing, and exposition. As always, the prompts from today's video are in the description under the YouTube version linked below.
18:34

DAY 7 of making a feature film with AI in 30 days. Today I show a clip of my film and discuss scene coverage, blocking, framing, and exposition. As always, the prompts from today's video are in the description under the YouTube version linked below.

Jenny Krakovsky

14,108 просмотров • 3 месяцев назад

I've posted my practice run of this talk on YouTube (link just below). This clip gives the core argument:
1:21

I've posted my practice run of this talk on YouTube (link just below). This clip gives the core argument:

Christopher Potts

11,793 просмотров • 8 месяцев назад

I gave a talk about why I'm skeptical of the argument for doom from alien AI motivations in "If Anyone Builds It, Everyone Dies." Video below and on YouTube, link to transcript and slides in thread. The talk is based on my recent essay on the topic, also linked in thread.
58:45

I gave a talk about why I'm skeptical of the argument for doom from alien AI motivations in "If Anyone Builds It, Everyone Dies." Video below and on YouTube, link to transcript and slides in thread. The talk is based on my recent essay on the topic, also linked in thread.

Joe Carlsmith

16,847 просмотров • 8 месяцев назад

What does the former co-lead of interpretability at Deepmind and now co-founder of Goodfire think about how to keep AI from becoming (his words) "a generally evil guy"? I got to speak with Tom McGrath at South Park Commons the day after OpenAI announced the Hugging Face hack. Good timing! (00:30) Why AI Interpretability Matters Now (02:15) When AI Reward Hacking Becomes Real (05:07) The Core Problem of Interpretability (07:23) Intentional Design: Steering What Models Learn (12:43) Neural Geometry: The Shapes Inside AI Models (22:29) Turning Interpretability Research Into a Product (25:28) How Research Changes Inside a Startup (32:48) Science Fiction, AI Risk, and the Future
35:40

What does the former co-lead of interpretability at Deepmind and now co-founder of Goodfire think about how to keep AI from becoming (his words) "a generally evil guy"? I got to speak with Tom McGrath at South Park Commons the day after OpenAI announced the Hugging Face hack. Good timing! (00:30) Why AI Interpretability Matters Now (02:15) When AI Reward Hacking Becomes Real (05:07) The Core Problem of Interpretability (07:23) Intentional Design: Steering What Models Learn (12:43) Neural Geometry: The Shapes Inside AI Models (22:29) Turning Interpretability Research Into a Product (25:28) How Research Changes Inside a Startup (32:48) Science Fiction, AI Risk, and the Future

Jonathan Brebner

12,234 просмотров • 15 дней назад

“The journey is part of the understanding.” As part of his guest edit, historian Tom Holland wants to talk about the impact AI has had on research. Writer Kate Mosse and Amol Rajan talk about why using AI as a research tool is not necessarily the best option.
1:22

“The journey is part of the understanding.” As part of his guest edit, historian Tom Holland wants to talk about the impact AI has had on research. Writer Kate Mosse and Amol Rajan talk about why using AI as a research tool is not necessarily the best option.

BBC Radio 4 Today

20,836 просмотров • 8 месяцев назад

Sharing a clip from my ‘Dhurandhar: The Revenge’ review, a film that drops pretence and makes no attempt to hide its acquiescence to power. YouTube link in tweet below ⬇️
0:29

Sharing a clip from my ‘Dhurandhar: The Revenge’ review, a film that drops pretence and makes no attempt to hide its acquiescence to power. YouTube link in tweet below ⬇️

Sucharita

37,779 просмотров • 5 месяцев назад

A clip from my conversation with Gary Oldman talking about Jackson Lamb. Link to the full podcast in the tweet below or in your pod app.
0:48

A clip from my conversation with Gary Oldman talking about Jackson Lamb. Link to the full podcast in the tweet below or in your pod app.

Barbican Station - A Slow Horses Podcast

10,046 просмотров • 1 год назад

DAY 19 of making a feature film with AI in 30 days. Today I talk about boring things like asset management. Follow my journey on my YouTube channel linked below!
6:21

DAY 19 of making a feature film with AI in 30 days. Today I talk about boring things like asset management. Follow my journey on my YouTube channel linked below!

Jenny Krakovsky

15,114 просмотров • 2 месяцев назад

Today we’re announcing a finding that breaks a core assumption in AI: that bigger models are harder to understand. We show the opposite. When interpretability is built into training, models become MORE understandable as they become more capable.
4:24

Today we’re announcing a finding that breaks a core assumption in AI: that bigger models are harder to understand. We show the opposite. When interpretability is built into training, models become MORE understandable as they become more capable.

Guide Labs

15,884 просмотров • 2 месяцев назад

This is a clip from a tourism documentary in Ireland by Callum A local from Belfast explains some of the local history. Link to full documentary below.
1:08

This is a clip from a tourism documentary in Ireland by Callum A local from Belfast explains some of the local history. Link to full documentary below.

Real News Éire

10,547 просмотров • 9 месяцев назад