Sam Altman: OpenAI, GPT-5, Sora, Board Saga, Elon Musk, Ilya, Power & AGI | Lex Fridman Podcast
ฝัง
- เผยแพร่เมื่อ 4 มิ.ย. 2024
- Sam Altman is the CEO of OpenAI, the company behind GPT-4, ChatGPT, Sora, and many other state-of-the-art AI technologies. Please support this podcast by checking out our sponsors:
- Cloaked: cloaked.com/lex and use code LexPod to get 25% off
- Shopify: shopify.com/lex to get $1 per month trial
- BetterHelp: betterhelp.com/lex to get 10% off
- ExpressVPN: expressvpn.com/lexpod to get 3 months free
TRANSCRIPT:
lexfridman.com/sam-altman-2-t...
EPISODE LINKS:
Sam's X: x.com/sama
Sam's Blog: blog.samaltman.com/
OpenAI's X: x.com/OpenAI
OpenAI's Website: openai.com
ChatGPT Website: chat.openai.com/
Sora Website: openai.com/sora
GPT-4 Website: openai.com/research/gpt-4
PODCAST INFO:
Podcast website: lexfridman.com/podcast
Apple Podcasts: apple.co/2lwqZIr
Spotify: spoti.fi/2nEwCF8
RSS: lexfridman.com/feed/podcast/
Full episodes playlist: • Lex Fridman Podcast
Clips playlist: • Lex Fridman Podcast Clips
OUTLINE:
0:00 - Introduction
1:05 - OpenAI board saga
18:31 - Ilya Sutskever
24:40 - Elon Musk lawsuit
34:32 - Sora
44:23 - GPT-4
55:32 - Memory & privacy
1:02:36 - Q*
1:06:12 - GPT-5
1:09:27 - $7 trillion of compute
1:17:35 - Google and Gemini
1:28:40 - Leap to GPT-5
1:32:24 - AGI
1:50:57 - Aliens
SOCIAL:
- Twitter: / lexfridman
- LinkedIn: / lexfridman
- Facebook: / lexfridman
- Instagram: / lexfridman
- Medium: / lexfridman
- Reddit: / lexfridman
- Support on Patreon: / lexfridman - วิทยาศาสตร์และเทคโนโลยี
![The Possibilities of AI [Entire Talk] - Sam Altman (OpenAI)](http://i.ytimg.com/vi/GLKoDkbS1Cg/mqdefault.jpg)
Here are the timestamps. Please check out our sponsors to support this podcast.
Transcript: lexfridman.com/sam-altman-2-transcript
0:00 - Introduction & sponsor mentions:
- Cloaked: cloaked.com/lex and use code LexPod to get 25% off
- Shopify: shopify.com/lex to get $1 per month trial
- BetterHelp: betterhelp.com/lex to get 10% off
- ExpressVPN: expressvpn.com/lexpod to get 3 months free
1:05 - OpenAI board saga
18:31 - Ilya Sutskever
24:40 - Elon Musk lawsuit
34:32 - Sora
44:23 - GPT-4
55:32 - Memory & privacy
1:02:36 - Q*
1:06:12 - GPT-5
1:09:27 - $7 trillion of compute
1:17:35 - Google and Gemini
1:28:40 - Leap to GPT-5
1:32:24 - AGI
1:50:57 - Aliens
💗
The fact that these two still think consciousness isn’t something unique to life is laughable! You will never build general intelligence! You don’t even know what consciousness is!
The fact that these two still think consciousness isn’t something unique to life is laughable! You will never build general intelligence! You don’t even know what consciousness is!
You forgot to timestamp the 132 ADs.
Do you think that in 2035 with the advent of AGI (Artificial General Intelligence) and ASI (Super Artificial Intelligence) they will introduce universal basic income!? a subsidy!? there will be free will!? an era of abundance will begin!? the intellectual and motor capabilities of robots have already surpassed the average human and before 2035 there is talk of them surpassing those of any human and then again those of any human combined! regarding both body and mind! answer me please...
I think I speak for everybody when I say, get Nvidia's CEO Jensen Huang on the podcast!
The leatherman.
Hyperbole overload.
But I am afraid he might not talk a lot besides marketing material.
@@cbuchner1Lex asks very thoughtful and specific questions though, I have faith that Lex would be able to get a genuine conversation out of him
Nvidia is literally powering chatgpt
Don't know why but I get a bad vibe from this guy
He's the very definition of machiavellian, so yeah...
They’re all sociopathic narcissists in tech. Here’s to hoping it’s not “good game Earth we had some fun huh?” with regards to the advancements of AGI and more butterflies and rainbows 🤷♂️
Me too
He was fired from YC because of his evil and self centric character. There was also an attempt to fire him from openAI due to his reckless behaviour. He is a dangerous man
You're probably a reddit loser who doesn't know anything about AI. You probably hate Elon too.
I never thought I would say this but Lex looks more alive than his guest.
Lex asks straight forward questions off a paper and gets zero answers, His podcast is taking about 3% of daily energy to maintain
Feels like he’s portraying a humble, sensitive persona when he may actually be an egomaniac with unbelievable power at his disposal. Didn’t learn much from this interview, but still engaging and important stuff watching Altman’s communication strategies.
Yes! All I took from the entire board saga in the end: I was hopeful/positive for OpenAI before, now I'm scared and get a nefarious vibe I really don't like...
I very much agree. It seems like the main things we can learn from a podcast like this is how someone wants to be portrayed
Exactly what I felt, you got it to the point
Don't tear something down just because you don't understand it.
Do you actually understand why he makes you uncomfortable?
I think we may have to admit that we all learn far more from uncomfortable uncertainties than the pleasant echo chambers we understandably tend to like.
Not to be unkind or to root for drama but I'm certainly curious about the giant reaction this guy's uneasiness generated....
Might be something worth taking a look at there...
Examination of all parties (we viewers/commenters included) illuminates a lot about psychology and sociology. 😅
“Take me through the OpenAI board saga”. My boy Lex got straight to the point XD
Let's take everything with a grain of salt until this goes through the court and all records are public. "Regurgitate me the propaganda narrative you came up with to save face while employees rallied behind you so they could get more than a $2 billion payout from a not-for-profit [and let's avoid speaking of your sister's many years of allegations of you sexually assaulting her for years as a child - where you seem desperate to attempt to maintain the head position of an organization that can AI generate realistic CSAM]."
Too bad Sam's reply was a whole lot of nothing.
"Tell me how Blackrock totally didn't install stooges on your board of executives in order to secure powerful AI for their sentient killbots that will eventually come for us all"
And he said nothing.
@mattmyers2624 for someone saying to wait and see you've sure seemed to have made up your mind
Lex is the only channel I check the length of the video thinking: "only" 2 hours? ❤
My Overoverovergod
Try Dwarakesh too. Dude does a lot of research
Sammy has no time to waste.
1 hour at 2x speed 😉
Yeah nice seeing two Jewish robot talking
I love Lex poked from every angle trying to get Sam to talk about about the openAI drama, and all responded with deplomatic answers or redirections. But then he asked how Sam felt, his personal feeling through this whole drama, finally it cracked.
Poked every angle ? He plays both sides while always choosing power over his feelings . Lex doesn’t have his own thoughts and sam isn’t here to play friends anymore
What timestamp?
@@kaynewest6167 I think Lex does indeed have his own thoughts and Sam has a lot of focus on him worldwide so he has to be diplomatic and reserved in every answer he gives, anything he says can cause significant issues or fears at least, so he has to be reserved
@@kaynewest6167 stroke victim argument
Sam Altman would make a great politician! He talks a lot without actually saying anything.
as he should. a lot to learn for future business owners that want to succeed.
Sam is under a microscope by world leaders . would you be screaming your blueprints or having friendly conversations with interviewers ?
Would be great to see an up and coming interview with Ilya
what did he see?
would be great to see proof that Ilya is still alive at all
Week later of the ouster, Altman came out and Sutskever went in the humiliation cave.
With a tour over the nuclear facility
@@Morskoy915 That "doesn't exist".
Two hours of information withholding.
yes
Yeah at the moment all he's talking about is the company dynamics. It's quite boring.
oh the irony
Maybe because Lex spent the first 20 minutes badgering him about the OpenAI drama like a tabloid journalist
he was soft-balling the whole time though...@@chinmarin380
This IS AGI answering the questions.
Exactly
Hahaaaa
Elon Musk, Mark Zuckerberg, Sam Altman, Jensen Huang and Bill Gates in one room with Lex hosting the discussion in a round table discussion would be insane and totally relevant. This would be the ultimate podcast.
I would not want to watch that on a full stomach nor an empty one for that matter. It would make my guts churn.
@@doublevision5465don’t be so dramatic💀
+Elon Musk :)
Add Ilya Sutskever and some engineers, I wanna watch the brain behind all this technology taking shit. Not just business head motherfuckers, LOL.
Add Ilya Sutskever and some engineers, I wanna watch the brain behind all this technology talking sh*t. Not just business head m0th3fuc*3rs, LOL.
It’s nice to include robots like Sam Altman in the conversation
what about cyborgs like Zuckerberg?
This is just two AIs conversing with each other
@@geoms6263did you mean reptiloid?
He really does sound like a robot, doesn’t he?! 😂😂
I don't know what are you talking about i see two robots 👍🏻
Lex asked all the right questions, but I remember nothing from Altman's answers after the interview completes.
He clearly said we have great people working on it, and there's a big group of very smart AI researchers, and I'm actually very happy that we're doing that, because we need to think more about these things, and we don't want to overestimate the risks, and moving towards the future, we still have much more to take care of.
You, my boy, have a long way to go if you think those were all the right questions
All part of the plan
Sam isnt really that deep, he just happens to lead a top AI company. Discussions about other civilizations and simulations are always interesting but he was just dead quiet there. Elon and Andrej really answered those questions quite well with deep thought
It felt a lot that he can't say to much or don't want. Both about technical stuff and the drama. More about the vision of the future may have been easier to talk about.
Thank you Lex. For these long discords to be able to hear and SEE someone's intent
Thank you for this interview 🫶🏻
This dude speaks like such a politician
talks 2 hours but only moves air and doesn't say anything - scary
@@jochennel He actually said a lot , were you listening?
People need to take their time and form their thoughts, he's not running a Burger joint
@@esyjournalhe can say a lot, doesn’t mean it has substance. He said in a lot of words just in the beginning “AI/AGI is going to have a power struggle” without elaborating on it. Look more into Sam and the company lol.
@@Timbo360how would you like him to elaborate on that? There's not much more to say without a crystal ball
This guy gives “The character who is on everyone’s good side, but actually turns out to be the villain” vibes
I swear. Watch him closely he feels very emotionally dead and machivellian at the same time.
Dude has a prepper bunker full of guns and years worth of food, tells you all you need to know about how he sees the future playing out
Dead-on. Him talking about how "painful" the board situation was for 20 min was hilarious. "But there was a lot of love too"
It's like those amateurish horror stories on reddit about Skinwalkers imitating humans.
Kinda like book littlefinger
facts
Lex you have so transcended any interview of Altmann and any interview about AI. You patiently, empatheticly asked and then let him search internally for his answers, his thinking - good or bad. Thank you sir.
Thank you, Lex, for these interviews.
No question about sexual abuse of Annie Altman, ofcourse. What a cool guy this Sam Altman.
Never thought Sora would create a Lex's Podcast
we are so close to this being a reality…
@@frankc.astle214we're fked
My Overoverovergod
😂😂😂😂
I got exactly the same feeling, Lex's hand gestures, video quality. Perhaps perhaps not lol
The amount of vocal fry is truly Elizabeth Holmes level.
what is a vocal fry? i know only about french fry..
@@hyennussquatch4597 youtube it, bro
@@hyennussquatch4597It's a meme that associates vocal raspiness with insincerity
Kim Kardashian 😂
Yes. Really annoying lol
My AI Agent. So his endgame becomes clear. I understand social phobias and fear of deep connection but he just revealed that he wants or is motivated by wanting a personal AI agent that will know literally everything about him and accept and help him - a feeling he wants from people in his life but has not had. This is a frightening motive for creating a “system” that could hold so much harm to human society and life. I didn’t expect to be so alarmed when starting this podcast. Lex has been amazing in this interview.
Thank you very much for a good interview.
are all Sam answers missing Lex questions point, or is it just me?
This dude sounds like an NPC
Yes, he did it the last podcast as well! Typical CEO behavior but Sam Altman really does it a lotttt
Sam gives me manipulative liar vibes
After scrolling through quite a few comments I finally found this thread. I'm surprised more people aren't pointing this out. Yes, Altman gives CEO answers, but also... He's learned a "love" vibe too well, and I feel like he's using manipulative tactics constantly on top of that. I don't quite trust him. There are a couple short moments that I think are real, but there are more moments where it's hard to tell if he's honest at all... Certainly he reframes each question slightly to avoid much true detail...
When he said "I love Ilya" you could just sense that deep down he probably feels the exact opposite about him. People like Sam Altman are really really dangerous.
Lex has gotten more assertive and I'm all here for it. Especially because he's well-informed so it's not just like typical media grandstanding.
He wasn't very assertive with Netanyahu - I feel like his credibility in my eyes definitely took a hit with that choice of interview, he could have really asked some big questions but dropped the ball completely
Lex the robot's A I. Is learning to be more assertive. Soon he will go full Terminator.
Of course, giving a platform to a man committing genocide was hardly the best idea to begin with lol
True, except for his interactions with Elon Musk, where he turns into an excited puppy 😆
Fan boy, idolizing the biggest fan boy. Lex is just trying to make smart friends, not give us smart conversations 😂
1:03:13 - Qué es Q* y avances más graduales o abruptos, GPT-5.
1:28:38 - Mejoras de GPT-4 a GPT-5
0:44:23 - GPT-4 kind of sucks...
0:47:46 - ChatGPT como interfaz y momento clave.
1:00:30 - Pensar rápido, pensar lento.
Gracias Dot! Estuve en tu directo
merci
Many thanks. Gracias por un programa y una entrevista de lujo
How can someone so smart teach us so little after a 2 hour interview
By being very diplomatic and careful not to offend anyone.
@@pushkarbaral4064You mean,* not to inform, I hope.
yes @@imaspacecreature
he was a pure soul when he started, and now the corporation has corrupted his soul,
his mouth is tied. He spoke his heart , and served the world. Now he speaks his mind, and serves the devil. Probably oversimlifed and overstated. He is trying to save his soul. But the internal struggle is impossibly hard. @@imaspacecreature
@@pushkarbaral4064So that's what the liberal mantra "right not to be offended" gets you. Everyone afraid to say anything about anything, so that fragile minds don't get upset.
the vocal fry torments me
Aaaahhh I hate it so much😡
Sexually tormenting?
It makes me feel claustrophobic 😭
The fact that this is the only thing you take away from this video speaks a lot of your intelligence
I can't handle it, it's too annoying!
'Hello' will always be my first word in a prompt.
Haha same here lol. I say “thank you “ too 😂
@@stevrgrs exactly, if we don't teach AI good manners... SKYNET
@@stevrgrs y'all are weird ngl
Yeah, don't worry sweet talking AI. Worry about not saying anything to the politicians using AI to screw you. Just recently, EU created an AI law where it makes certain AI illegal to use by anyone but the government. So prepare for the social credit system.
This is actually really important. To teach AI respect compassion empathy teamwork. It matters.
Love watching these while working on a fun project in the background.
Straight up
Lex is spicy in this interview! I love the new pressure he’s putting on Sam to answer his questions. Hoping to more of this in future episodes.
he's such an elon cuck and i'm a fan of Lex
He's not really challenging. He's joking. Wish it was a bit harder questions.
there was no pressure, cmon :D
Yes being elons side
@@dusanbosnjakovic6588 what kind of hard questions?
Everytime I watch Altman I trust him less and less
it’s funny because every new interview he has gotten more powerful, either he is becoming more and more smug or more and more corrupted by such power.
pointless interview and you can tell Lex knows it with his sarcastic jokes late on. Altman answers in CEO form. There was some better responses late on but the first hour of the interview was just a waste of time.
I hope he won't see your comment, cuz that would just be devastating for him
Everytime I read a comment like this, I think most people are shitty
they are mostly talking about nuclear fission (what we have today) and not fusion, correction for the subtitles! After mentioning fusion at 1:11:45
I took nothing away from this interview
@@gunzor8717Rogans feed is your average edgy 14 year olds I bet, gore, war, fights, cop shootings, and animal attacks
Then it's not your place i guess
it's okay you couldn't keep up, go watch some tik toks or something
Meh, it's TH-cam. You couldn't fit more attention whores in if you packed them sideways.
that’s impressive 🗿
His voice is like an itch on my brain I can't scratch
On the cusp of AGI, the doorsteps to absolute power stands Sa-ah-ah-am.
Seriously though, we are really going to let this vocal fry communicating robot take over the world?
@@TheIronpusher🤡
😂 Not sure if thats a compliment or insult🤣
@@darylallen2485what about that would be a compliment
It’s called vocal fry I think its common in San Francisco people
Sometimes sam's pauses are so quite I keep checking my wifi connection
Fascinating. Lot of missed opportunities to hear answers to some important questions the host asked. I’m curious. Some deflections. Interesting clues. Thank you.
In the clip (57 Min) he totally avoided the question of tranparency and data privacy of user and started talking about his firing and tried to appeal emotionally. This guy is shrewd. how long is AWM19V opened to us?
What??
They are both jewish. They see themselves as of a higher caste with in-built nepotism of extreme proportions.
Bot
They are both Zionists. What did you expect from them?
Because it doesnt matter everything you use collects your info
1:09:22
“You're not the man you used to be, Sam."
"Very different job now than what I used to have."
I watched this part so many times and it still sends chills down my back. I listened to the 2 podcasts with Sam so many times and examined them in detail.
In the first podcast, Sam's stance was "Look at the amount of pettiness among the "experts" when we started, now look where we are. We are going fast and going strong. Great accomplishments, huge milestones, bright future ahead." He was the guy defying all odds and proving everyone wrong by making something remarkable.
The future was unknown, and in a dark cave, there was light.
Now things have completely changed. By moving towards that light it becomes clear that AGI is not an IF but WHEN, and as things move forward at exponential speeds, we are racing toward that light we see, kicking and pulling on the people around us, and what's waiting for us at the destination? Would we run off a cliff without being able to stop?
The level of certainty in his face and voice just 11 months ago that OpenAI would either make AGI or "not make AGI but still make humans super great", and the level of uncertainty we see today that "AGIs are coming any second and the world is not ready for it", is insane.
What was remarkable between then and now to me was his wisdom; his going on record to say essentially he'd at least hope to set aside his own need/desire for power in exchange for saftely. How fortuante for the world that someone like him is at the helm. I love Elon but its rather obvious that we cannot trust him with irreversible power buttons.
Thank you for pointing out the timestamp
I rewatched it several times. Got chills as well
Really make you think how much one can be changed by many. It'll be an interesting movie to watch in 10 years
I don't understand, so you mean he became less certain about AGI?
@@VejmR I take it to mean more certain it's on the horizon, less certain the uses will be worth the risk. I suspect Sam has a much better understanding of how even people he would have innately trusted can be swayed by the stress of power. That doesn't extrapolate optimistically to the development or potential uses of a tool as powerful as AGI.
@@VejmR To put it simply, he has become certain that AGI will be here, most likely very soon, and now the work has changed from "Can we maybe create AGI?" to "What do we do with AGI to not ruin the world."
It's the same with nukes. "How do we build a nuke?" is a scientific problem. He was Oppenheimer. After nukes were made it becomes a political game: "How do we leverage our nukes to gain advantage in world politics?"
His identity is shifting or has already shifted from being a research scientist to a politician. That's my understanding of this line he said.
Edit: I can't spell. Removed an extra "s".
I think this is one of the best podcast endings I have ever seen @lexfridman ! Super excited to hear about your journey to the amazon! and your trip figuratively and literally lol.
1:33:50 Yes it did change the world. In the future, November 30, 2022, will be looked at as an inflection point. It is when there was a leveling of the world of intelligence. I was 67 and suddenly i can develop software applications, do PhD level research. Ask the deepest questions of my heart, and on and on.
Sure it has many limitations but before I could even complain, new models with more capabilities were released every month. Just when i knew how to get a certain task accomplished without error (hallucinations), some newer plug in, the gpt comes along and that's no longer a problem.
And just over a year later, we have text to video (Sora)!
Absolutely spectacular growth and advancement. I've never been more excited to be alive. 🔥
thanks for sharing your perspective :) i'm disabled and feel the same way 🥰❤️🌈
any chance you will get Ilya again? if possible would be great to hear both sides of the saga
While I'd love to hear from him, you generally have to hear one side of the saga before you can hear both sides. And so far, we've heard no sides, just some emotional responses to it.
I think Ilya is hanging out somewhere with Kate Middleton
We need to start searching on nuclear facilities
I find Ilya to be the most interesting and informative AI persona, definitely want to see more of him.
Tbh I don't want to hear the saga from Ilya, I want to hear his insights and wisdom on AI.
I'm effectively getting real altruistic Sam Bankman-Fried vibes off this guy.
@TheThriftShopSampler The chances altruism plays any role in what he does is essentially nil.
If I'm recalling correctly Sam is or was part of the real altruist movement
100%
@@tristinyaeger9222 that's what you do when you want to hide in plain sight
We need Elon, Ilya, Max Tegmark and Sam in one room. Lex as moderator..Please.
38:00 I love this part. The negotiation.
Evening lads, well, i've watched this twice now...with friends and family. Slightly concerned, as one should be, but this is indeed the world we live in. Thank you Lex for the platforme.🇫🇷🥂
Sam Altman gives me super villain vibes
Wait until you hear about the theory that he is actually trying to take over the world
Yeah, dude is creepy...
I think it's just his generation. Uninspired, boring and just out there fulfilling the captain of industries dream. Automating economic output at all costs. Nothing really exciting or imaginative about it. Maybe the machine will produce something exciting... almost like a script being followed.
@@paulm3969 that’s a pretty profound take. Thank you.
That's just because he's so much smarter than you.
Sam: Our goal is not to have shock updates to the world.
Lex: What is Qstar?
Sam: We're not ready to talk about that...
I love all your content, especially what is focused on AI. I'd love to listen to you interview William Gibson to discuss how realistic the plot of his novel Neuromancer may seem now. Will AI constructs compete with each other?
1:06:30 "what's the way we release this thing", makes me think he refers to making Sam give honest first hand answers. I adore the dynamic that's happening. It really is about getting Sam to open up. But of course, it's hard.
Lex first question "Take me through the OpenAI board saga". *Sam Altman* - I had the feels and stuff. Way to answer the question....
EXACTLY! He talks and talks but conveys nothing! = nefarious things happening.
Sounds like Mark Zuckerberg when being asked questions from congress
@@MrSeanman30You mean like the question what data types they store?
Or when if a guy walks across the room with an iPhone, does Google see that? Oh wait, that was Sundar.
yeah, i'd be as confused as they were.
“For a second I felt like a human being, and I never want to feel that again”
The most interesting thing about this interview is how many people think it's a great interview.
that's every Lex podcast
Not a lot
What problem did you have with the interview?
That's interesting. I have to think about that.
a lot - like… more than 3?
Nice to see so many great pocker players here who can read the people mind just by watching a video interviews with them. Seriously, you guys are so good at distinguishing people with good intentions from the opposite?
Yes, thankfully people have this ability. It protects them from being taken in by charlatans like Trump and this guy.
I love that he wants the board to be answerable to the whole world.
Extremely interesting. Thank you Lex
Hola Marti 😅
Is fascinating
@@numbercruncher6242 You mean, he says a lot, but tells nothing.
I clicked on this so fast I almost lost a finger 😂
y adiosss
In the clip (57 Min) he totally avoided the question of tranparency and data privacy of user and started talking about his firing and tried to appeal emotionally. This guy is shrewd.
I noted that he didn't answer the question of what ACTUALLY happened at OpenAI. Still don't know why they fired him, only how hurt he was by it
Yeah I found "User choice" to be very broad.
This was a huge red flag to me. Not only was it an issue I wanted to hear discussed (its huge in the public space right now) but the way he shifted the attention to his personal pain? It was the exact wrong moment to invite empathy. Nope! Sorry Sam burn out isn’t a get out of jail free answer to ethical concerns.
@@aprilloft100% spot on!
Thank you Lex, that invite him I get many something from him and you that make question. PS. I never know that ... thinking about Capital words
Damn it! I thought this was an interview with The Disco Biscuits original drummer. Maybe get him on sometime Lex
Sam Altman -*TRUST ME BRO*
Too bad you can't see the nuances
@@KoryogdenToo bad you can't see the obvious.
I had to check my phone every time Sam pauses thinking an Ad is popping up, but bro is actually thinking not to mess things up 😂
Same my airpods have been playing up and turning off and I keep turning them off trying to turn them on
Yeah, unfortunately he is anxious. And sadly, people are like suspicious because they can't comprehend human emotion
THIS GUY IS A WEASAL.
Well. He runs the hottest shiet in the tech industry. There's big money from big players involved.. i'd be veeeery careful too sitting at the top of a 100 billion $ company 😂. You too
@blaaaaaaaaahify these randos cant understand, theyre just content consumers.
Hi there, would you (or do you plan to) interview Emad Mostaque (formerly) of Stability AI. I would just love you two exchange ideas and you to pick the brain of his.))) Thank you for your work, Lex! [deep bows]
I think Sam has trapped Ilya's consciousness in a wall mounted robot face so he can be best friends with him forever
"I personally hate lawsuits" what a great insight
shill
ofc. They use all the data available on the internet that's not theirs. Stronger case to put forward on podcasts etc. that you don't like lawsuits.
sam seems very untrusting of lex so far lol. (I'm only 30 mins in so hopefully changes)
really wish there were more altman interviews elsewhere so i don't have to listen to lex
@@rasnwell given that Lex is Elon’s friend…that checks out with Sam being sued by Lex’s friend
The master of vocal-fry. Could be a cool drinking game: Everytime Sam pronounces a syllable WITHOUT vocal-fry, everybody has to drink a shot. :D
I can’t unhear it now 😭
fucking hell that fake voice. Unbearable.
I hate it.
Please I'm so thirsty
Vocal fry is a tactic used to feign humility while screwing over the intended audience.
Civilizations have certain milestones that change the direction of humanity, this is one of them. I have been in AI for 30+ years and this time the impact is real as can be seen by the number of people working on it and doing incredible things with it. This is just the start. You both are doing great in explaining it to the word.
This was a very enjoyable interview. Thank you Sam, and Lex. I see a lot of people with fear in the comments, and I feel like they attribute that where it probably doesn't need to be.
I never thought I’d see Lex interview the founder of Skynet.
Lex Luthor vs Miles Dyson
He isn’t the founder though. He was just a PR guy
At least we got a good Lex interview before he handed the world unwittingly to the demon he has no idea exist ruling the world. Sad dystopia he is rushing us off to as Bill Gates 2.0, surprising Altman sees no issues with Bills insane control of humanity, down to owning patent rights for organic seeds.
. Sam is lead developer and picked the team
He is Skynet's marketing guy, neither founder nor tech guy.
Lovely to see Sam from ClosedAI
Interesting how he avoided the question about why Ilya thought he needed to go, isn't it?
Am I the only person who feels sick after watching this interview rather than excited for the future?
@lexfridman no wayyyyy - I plan to go into the Amazon next year for various weeks myself. Hopefully to work with animals. Crazy cannot wait to hear how you go!
Lex is a force of nature. Not only does he get the most influential guests but it also seems like the guests seek him out and it is just as important for them to be on with Lex. Amazing journey, congratulations Lex, you are awesome!
Thanks, other O'Keefe.
Of course, after all Lex does not ask any difficult or important questions. The guests basically promote themselves, and I also feel Lex prepared very poorly for this interview. No in-detail questions, just layups and easy way outs...not a satisfying and value added experience. I learned nothing new that I did not already know.
The only weakness Lex has, I'd agree. His style is incredibly non-combative@@bbmatfreddy
@@bbmatfreddy He's not a reporter. His aim isn't to dig up dirt on his guest, it's to let them speak and give insights into their own knowledge/wisdom. If you think his questions were too surface level, then that's a valid argument.
@@Astra2 he takes on the function as investigative journalist. In that he does a pretty poor job. Not sure what else his purpose in such interview would be. Altman could as well have just used a Sora bot to ask easy questions that make him come out on top of everything.
For example, why did Lex not press harder on the opensource issue. Letting Altman simply say "I don't even know what this lawsuit is about" is as disingenuous and dishonest as it gets. OpenAI not being open at all. Letting tons of outsiders contribute for free, for Elon to invest millions, and then shutting the whole thing down and distributing unheard sums to friends and family, aka employees. Really? Lex had nothing to ask about that? Or the issue of Sora being trained on unlicensed content. Whatever one's position on this may be, it's an important issue worth many upcoming lawsuits. I stand by my stance, one of Lex's worst conversations and interviews and a great chance wasted to get to the bottom of many very important topics.
We need Elon, Lex, Sam and Zuckerberg in the same room next episode. we demand it
😮
@@SuperYogagirlElon is much cooler than all of them combined. You must be very woke to harbor such intense hatred towards him.
Hahaha that would be so awesome omg
That will be an Alien slasher movie.
@@SuperYogagirl not everyone has a weird hate boner for Elon like you, Elon would be my number 1 guest on that list of names. By far.
- Loved the shutdown on the analogy at 12:35
- Another gem "Look I understand you want to spend the first 20 minutes talking about this dramatic weekend" 18:05
- 42:03 Reframing how to think about AI job replacements. Think instead about 5 second/minute/hour/day/week jobs. How many of those can AI replace? Typical job is a mixture of these.
- 54:25 Journalist's usage of ChatGPT
- 1:05:00 OpenAI is trying to do iterative releases so that the public becomes acclimated. Uncomfortable that external parties seem to think there are "leaps"
- 1:11:00 Compute doesn't have a defined market saturation (like physical goods). Better thought of as intelligence. Would always prefer more, but at different prices, different amounts can be applied. More compute is going to be an enabler.
- Lex saying multiple times "I'm not sure if you know humans, but ...." Like he's speaking to an alien. An example: 1:12:25
- 1:16:45 Elon's pettiness. Lex (probably realizing whatever he says next will be clipped says "Elon is a friend, beautiful human being, one of the most important...")
- 1:32:30 Really good point on AGI. Kind of a meaningless bar that no one will ever agree on. Focus instead on specific tasking
I don't understand this comment section. A lot of it is valid criticism, a lot just hate, but I think Sam did a good job answering. It was very interesting. Things have changed since last time.
I’m so glad Lex brought up the “What did you see Ilya!?” meme 😂
can someone explain me this. for some reason i dont get it after seeing it a million times
And the answer he got for it was verbal tossed salad.
I have bad vibes about Sam. Just very calculated about certain questions, I smell some dishonesty.
Also seems he calls all the shots, is above the board and chairman.
@@MacroAcc the meme suggests Ilya saw AGI or something worrying and this caused him to vote to kick out Sam due to the gravity of what he saw
@@MacroAccwhen altman got kicked out by the board, ilya sutskever who was a founding member and is the one driving most of the technical aspects of openai voted to kick altman out. The reason for that is unknown, which is what caused the meme. No one knows what happened at openai and what caused them to kick altman out, hence “what did ilya see (to make him want to kick altman out)”
Lex been having a good variety of interviews this year love it
Bro its crazy, so many good/interesting/extremely relevant guests! Just love his podcasts
id just like to say that both definitions of AGI by Sam are both implementations of the agi not the underlying code/software
MOD mixture of depths architecture assigns different amount of compute to different tokens. It routes simpler tokens (commas , accents ), so test they don't have to go through all the transformer layers and spend the compute
If you listen closely at 1:11:56 when Sam finishes saying Nuclear Fusion, his robotic parts make a glitchy sound
🤣
🤣🤣🤣
😂😂😂
Lex is basically a really awesome prompt machine! Thanks for this interview.
The Q-star thing is really interesting here. The fact that OpenAI is so quiet about it says a lot about it as Lex pointed out, I think that Q-star is indeed a new form of fine-tuning of GPT whose performance is so astounding that even OpenAI can not explain.
Awesome interview.
Thanks for asking most of the questions we put to you on LinkedIn Lex.
Amazing podcast 👏
Yes but Sam didnt answer to them. What a useless Interview.
This conversation, besides Sam’s comments on the board movements, largely mimic that which took place last year. I think these models have exceeded Sam’s technical depth. And I think the closed source, for profit business plan, Open AI has adapted to prevents lower level personnel from speaking to the challenges they are dealing with day to day, that the greater community might enjoy and derive real value from. This felt so high level that I could conceive it being synthetically generated.
Well said. A lot of comments noting they don't trust Sam/get bad "vibes" from him. Really all I'm picking up on are very calculated, non-inflammatory responses amounting to very little.
This is exactly how I felt. Initially I felt like Lex was trying to be more technical but the guy sounded like a software salesman who doesn’t want to get deeper. Had to google his background again cause I was very surprised
@@DeadRyGuy and to a certain extent you have to expect corporate speak from a guy in his position. Because the valuation correlates with his sentiments. But this product category is far more nuanced, if I was a shareholder I’d expect a lot more technical depth.
@@emmersonzhaime1998 Lex has interviewed Open AI’s real CTO Ilya Sutskever, albeit years ago, and years before they even had a product. I have no idea who this Mira Murati is claiming to be CTO but in interviews she’s somehow more inept than Sam. I believe she’s likely a token figurehead. The interview with Ilya was adequately technical, but he’s been radio silent since, almost as if he’s being intentionally silenced. I can’t make sense of much of this company, but as someone in the space, I am really curious to see how this plays out for them.
@@usefulprogrammer9880ilya is definitely hostage in that nuclear bunker 😂😂
All his answers were extremely vague and every possible edge was perfectly rounded
Sam, OpenAI's LLM has dramatically effected my life and my capabilities for the positive. Thank you for this most powerful tool.
Hey Sora, create a video with Lex Fridman interviewing Sam Altman for about 2 hours talking about AI and the state of the technology in society.
Yeah and then break in and post it from Lex' account to confuse everybody? Get real
Wow Sora's getting good, hey make sure the NPC says very little while speaking a lot.
@@Neptunade someone can’t take a joke. Sheeesh
Now have Ilya on for round 2
lol , sometime people make these requests as if all these guys guys are just lining up to be in the podcast and Lex just picks them lol
@TheHeavenman88 Yeah but if enough people push for it, the more likely Lex will do the same. Then it's up to the guest.
Hello
This is the key question that no one is asking St. Altman
A few months ago you mentioned that you were present when something was discovered that pushed back the veil of ignorance. The question is: Was that discovery 'Sora'? If the answer is no, then the anticipation for what it might be is even greater.❤❤❤❤
Thank you so much lex❤❤
These recent episodes have been so great, I wasn't able keep up with the high amount of podcasts you released before.
But the last couple of weeks I have been enjoying the heck out of every released podcast, thanks Lex for keeping to improve your process and make ever podcast better than the one before
Thank you for releasing in the middle of my mid term exams now I can't skip
Womp womp
You'll probably learn more out of this 1.5 hours than any formal education.
One advice from drop out student.
Don't waste that time. If you have opportunity to study what you like then study hard.
Easy to say.... I know... but
DO IT!
Focus on what is important. I believe in you :)
come back to this shiny thing after your exams are done
For the second time I faced an idea that ChatGPT could be a reading partner. Lex mentioned this here: 53:03
The question is how does it work?
Tell it what you read and what you thought. Ask it what it thinks
He talked about two hours without saying anything. Just did a slalom maneuver between Lex's sharp questions
- Are you training Sora on other peoples' IP?
- I think people who created valuable content should be compensated
- ok, thank you for answering
This is what this conversation looks like in a short form. Thank you Lex, very insightful
i dont care if it is trained on other peoples IP
@@doyourownresearch7297say that to me when your face is being used for "bank robbing attempt 176". Allowing it to use your most private data in something that can so actively harm you is a terrible precedent. You are inviting the most powerful digital panopticon.
@@doyourownresearch7297 You don't care about theft or intellectual property. Cool.
@@doyourownresearch7297guess you don't plan to create anything of value in your lifetime then
If the data and models etc. were all open source and not for profit I think it would be fine.
He never failed to talk down about Elon whenever there's a chance
Have you seen on Twitter how Elon attacks OpenAI every day?
@4l3dx have you seen how much money Elon funded in the early days?
@@Mopoppy Oh, right, so they should just stay quiet and take the attacks 😌
Elon is annoying as hell and needs to be checked. People need to stop worshipping that guy just like they worshipped Jobs. He says the most idiotic things and treats his own workers like trash.
Elon must is a doosh! If you can’t see that you are blind.
Great talk, thank you.
Most hilarious part is Sam *somberly* recalling when Elon tweeted a silver medal at Jeff Bezos after surpassing him as the richest person in the world 🤣