Dario Amodei is the CEO of Anthropic, the company that created Claude. Amanda Askell is an AI researcher working on Claude’s character and personality. Chris Olah is an AI researcher working on mechanistic interpretability.
Thank you for listening ❤ Check out our sponsors: https://lexfridman.com/sponsors/ep452-sb
See below for timestamps, transcript, and to give feedback, submit questions, contact Lex, etc.
*Transcript:*
https://lexfridman.com/dario-amodei-transcript
*CONTACT LEX:*
*Feedback* – give feedback to Lex: https://lexfridman.com/survey
*AMA* – submit questions, videos or call-in: https://lexfridman.com/ama
*Hiring* – join our team: https://lexfridman.com/hiring
*Other* – other ways to get in touch: https://lexfridman.com/contact
*EPISODE LINKS:*
Claude: https://claude.ai
Anthropic’s X: https://x.com/AnthropicAI
Anthropic’s Website: https://anthropic.com
Dario’s X: https://x.com/DarioAmodei
Dario’s Website: https://darioamodei.com
Machines of Loving Grace (Essay): https://darioamodei.com/machines-of-loving-grace
Chris’s X: https://x.com/ch402
Chris’s Blog: https://colah.github.io
Amanda’s X: https://x.com/AmandaAskell
Amanda’s Website: https://askell.io
*SPONSORS:*
To support this podcast, check out our sponsors & get discounts:
*Encord:* AI tooling for annotation & data management.
Go to https://lexfridman.com/s/encord-ep452-sb
*Notion:* Note-taking and team collaboration.
Go to https://lexfridman.com/s/notion-ep452-sb
*Shopify:* Sell stuff online.
Go to https://lexfridman.com/s/shopify-ep452-sb
*BetterHelp:* Online therapy and counseling.
Go to https://lexfridman.com/s/betterhelp-ep452-sb
*LMNT:* Zero-sugar electrolyte drink mix.
Go to https://lexfridman.com/s/lmnt-ep452-sb
*OUTLINE:*
0:00 – Introduction
3:14 – Scaling laws
12:20 – Limits of LLM scaling
20:45 – Competition with OpenAI, Google, xAI, Meta
26:08 – Claude
29:44 – Opus 3.5
34:30 – Sonnet 3.5
37:50 – Claude 4.0
42:02 – Criticism of Claude
54:49 – AI Safety Levels
1:05:37 – ASL-3 and ASL-4
1:09:40 – Computer use
1:19:35 – Government regulation of AI
1:38:24 – Hiring a great team
1:47:14 – Post-training
1:52:39 – Constitutional AI
1:58:05 – Machines of Loving Grace
2:17:11 – AGI timeline
2:29:46 – Programming
2:36:46 – Meaning of life
2:42:53 – Amanda Askell – Philosophy
2:45:21 – Programming advice for non-technical people
2:49:09 – Talking to Claude
3:05:41 – Prompt engineering
3:14:15 – Post-training
3:18:54 – Constitutional AI
3:23:48 – System prompts
3:29:54 – Is Claude getting dumber?
3:41:56 – Character training
3:42:56 – Nature of truth
3:47:32 – Optimal rate of failure
3:54:43 – AI consciousness
4:09:14 – AGI
4:17:52 – Chris Olah – Mechanistic Interpretability
4:22:44 – Features, Circuits, Universality
4:40:17 – Superposition
4:51:16 – Monosemanticity
4:58:08 – Scaling Monosemanticity
5:06:56 – Macroscopic behavior of neural networks
5:11:50 – Beauty of neural networks
*PODCAST LINKS:*
– Podcast Website: https://lexfridman.com/podcast
– Apple Podcasts: https://apple.co/2lwqZIr
– Spotify: https://spoti.fi/2nEwCF8
– RSS: https://lexfridman.com/feed/podcast/
– Podcast Playlist: https://www.youtube.com/playlist?list=PLrAXtmErZgOdP_8GztsuKi9nrraNbKKp4
– Clips Channel: https://www.youtube.com/lexclips
*SOCIAL LINKS:*
– X: https://x.com/lexfridman
– Instagram: https://instagram.com/lexfridman
– TikTok: https://tiktok.com/@lexfridman
– LinkedIn: https://linkedin.com/in/lexfridman
– Facebook: https://facebook.com/lexfridman
– Patreon: https://patreon.com/lexfridman
– Telegram: https://t.me/lexfridman
– Reddit: https://reddit.com/r/lexfridman
source
Thank you for listening ❤ Check out our sponsors: https://lexfridman.com/sponsors/ep452-sa
See below for timestamps, transcript, and to give feedback, submit questions, contact Lex, etc.
0:00 – Introduction
3:14 – Scaling laws
12:20 – Limits of LLM scaling
20:45 – Competition with OpenAI, Google, xAI, Meta
26:08 – Claude
29:44 – Opus 3.5
34:30 – Sonnet 3.5
37:50 – Claude 4.0
42:02 – Criticism of Claude
54:49 – AI Safety Levels
1:05:37 – ASL-3 and ASL-4
1:09:40 – Computer use
1:19:35 – Government regulation of AI
1:38:24 – Hiring a great team
1:47:14 – Post-training
1:52:39 – Constitutional AI
1:58:05 – Machines of Loving Grace
2:17:11 – AGI timeline
2:29:46 – Programming
2:36:46 – Meaning of life
2:42:53 – Amanda Askell – Philosophy
2:45:21 – Programming advice for non-technical people
2:49:09 – Talking to Claude
3:05:41 – Prompt engineering
3:14:15 – Post-training
3:18:54 – Constitutional AI
3:23:48 – System prompts
3:29:54 – Is Claude getting dumber?
3:41:56 – Character training
3:42:56 – Nature of truth
3:47:32 – Optimal rate of failure
3:54:43 – AI consciousness
4:09:14 – AGI
4:17:52 – Chris Olah – Mechanistic Interpretability
4:22:44 – Features, Circuits, Universality
4:40:17 – Superposition
4:51:16 – Monosemanticity
4:58:08 – Scaling Monosemanticity
5:06:56 – Macroscopic behavior of neural networks
5:11:50 – Beauty of neural networks
Transcript:
https://lexfridman.com/dario-amodei-transcript
CONTACT LEX:
Feedback – give feedback to Lex: https://lexfridman.com/survey
AMA – submit questions, videos or call-in: https://lexfridman.com/ama
Hiring – join our team: https://lexfridman.com/hiring
Other – other ways to get in touch: https://lexfridman.com/contact
EPISODE LINKS:
Claude: https://claude.ai
Anthropic's X: https://x.com/AnthropicAI
Anthropic's Website: https://anthropic.com
Dario's X: https://x.com/DarioAmodei
Dario's Website: https://darioamodei.com
Machines of Loving Grace (Essay): https://darioamodei.com/machines-of-loving-grace
Chris's X: https://x.com/ch402
Chris's Blog: https://colah.github.io
Amanda's X: https://x.com/AmandaAskell
Amanda's Website: https://askell.io
SPONSORS:
To support this podcast, check out our sponsors & get discounts:
Encord: AI tooling for annotation & data management.
Go to https://lexfridman.com/s/encord-ep452-sa
Notion: Note-taking and team collaboration.
Go to https://lexfridman.com/s/notion-ep452-sa
Shopify: Sell stuff online.
Go to https://lexfridman.com/s/shopify-ep452-sa
BetterHelp: Online therapy and counseling.
Go to https://lexfridman.com/s/betterhelp-ep452-sa
LMNT: Zero-sugar electrolyte drink mix.
Go to https://lexfridman.com/s/lmnt-ep452-sa
That last hour was the most information-dense hour I've listened to in a while.
I like this guy better that S. Altman.
the reason why a model feels dumber after a while is simple: we are used to interact with humans who have the ability to learn and adapt. and we forget that models dont do that, they always stay the same. so its not that models get dumber, its that normally humans get smarter or more aligned.
dope
their ai safety framework seems great!
I like to listen to a CEO talked passionate and details about his product like Dario. Don’t get the same feeling with sam, he feels more like a manager
Why not similate the roughly architecture of the brain? There are all sort of zones in the brain that do different things. Just a creative idea 😉
He talks like Elon Musk… "it, it , it …" "ah, ah, ah…"
Love it Learning so much – So many perscpetive making comp[any's so human I love it 🙂
Claude is way ahead in humanlike thinking and assistance in my opinion.
It has moved me personally a number of times.
Claude is way ahead in humanlike thinking and assistance in my opinion.
It has moved me personally a number of times.
how can someone just talk for 5hrs. ?
Bro doesn't have to shout to lex
Lmao at Dario thinking "brainstorming and creative writing" need Opus but "taxes and legal contracts" only need Haiku.
So when will androids dream electric sheep
AGI is decades away if not longer. Anyone who knows how machine learning works under the hood knows this. Anyone who says otherwise is blowing sunshine up people's ass.
This guy looks like Matthew Rhys from The Americans! 😂
I do not think there will ever be a lack of data. Humans are not using AI models for free. Humans are active participants in the building of larger, stronger and smarter AI models. Humans are providing data, private data, on a daily basis to the leading AI companies. As long as we keep using these AI systems, data will always be abundant. There is never something for nothing. We might run out of public data, but we will never run out of thought, thoughts that we chose everyday to hand over to these AI systems.
The issue with LLM/AI is the way it's being rolled out to investors. You gentleman are lying to investors when you don't bring along the progression charts and dutifully explain to them that you are at the edge now, you are investing in fantasy land with the hope that the roulette wheel stops on their # and they win a trillion $ before Palantir decides to nuke a few cities with its genius algorithms that so far have been targeting kids in Palestine. ChatGPT is a fancy toy, but a toy that will forever be 15-20% broken will not one day wake up and start generating human emotion and insight.
Lex… you’re a GOAT!
4:39:44
But I think there's also a lot of value in just being like, you know, I'm going to essentially assume,
I'm gonna condition on this problem being possible or this being broadly the right approach,
and I'm just gonna go and assume that for a while and go and work within that and push really hard on it.
And, you know, society has lots of people doing that for different things. That's actually really useful
in terms of going and getting to, you know, either really ruling things out, right?
We can be like, well, you know, that didn't work and we know that somebody tried hard.
Or going in and getting to something that it does teach us something about the world.
He looks like Dan Melcher from HBO Silicon valley, if they make a movie about this guy that's the actor to give the role
Imagine if somehow AI could be used on the stock market.
Алексей, в той части, где ваша собеседница стала рассказывать про общение с claude, я вспомнила историю из wsj про одного парня. он создал чат бот replika и назвал его «грушенькой». в честь героини романа «Братья Карамазовы». признавайтесь, это были вы?…
я недавно решила перечитать это произведение ради любопытства. в очередной раз убедилась. Достоевский – не мой автор. чтобы отойти от Достоевского, мне пришлось примерно 1,5 месяца "прикладывать к голове" книги Джейн Остин и Толстого.
а если серьезно. я сейчас много работаю с танцем, движением и телом. как режиссер и как танцор. я осознала одну простую вещь. все эти суррогаты в виде социальных сетей, чат ботов и ИИ никогда не заменят базового бытового взаимодействия между людьми. простого касания взглядами, пальцами. и тех ощущений, которые дарит это касание. суррогаты – это путь в никуда. человеку нужен человек точка From Russia With Love!❤
p.s. я не успела досмотреть до конца. но мне безумно понравился момент, когда в кадре появились лучики солнца, которые подали вокруг вашей собеседницы.