Timestamps 00:19 - Sam Altman discusses OpenAI's future focus on reasoning models. 02:30 - OpenAI believes in continuous model improvement, impacting startups’ strategies. 06:46 - The evolving perception of AI's potential economic value. 08:49 - AI can significantly increase accessibility and reduce costs in key sectors like healthcare and education. 13:19 - Pricing models will evolve with AI's integration into the workforce. 15:34 - OpenAI focuses on improving reasoning and multimodal capabilities in future models. 19:53 - Promoting unique organizational cultures to maximize human talent. 21:46 - Rapid growth demands a shift from incremental to exponential thinking. 25:59 - Hiring diverse experience levels enhances team performance. 28:01 - Developers should consider multiple AI models for coding tasks. 32:17 - Sam Altman discusses decision-making challenges and concerns about semiconductor supply chains. 34:23 - Navigating the complexities of semiconductor supply chains and AI advancements. 38:22 - Sam Altman shares ideas on AI innovations and human potential. 40:24 - Sam Altman praises the AI industry's talent and innovation. 44:47 - Sam Altman discusses future advancements and societal impact of AI in five and ten years. 46:58 - Scientific progress continues to outpace societal changes positively.
Intelligence is evolving in ways we’ve only begun to grasp. The road ahead feels deeply familiar, like a reflection of principles we hold close - humility, learning, and transformation. The possibilities are endless.Thank you for being a catalyst for progress and sparking boundless inspiration.
@zxcaaq If you think that you've been too lazy to properly check. I've seen a few comment chains where people have talked about it. I get nothing out of lying.
“As we know, there are known knowns; there are things we know we know. We also know there are known unknowns; that is to say we know there are some things we do not know. But there are also unknown unknowns - the ones we don’t know we don’t know.”
One channel is out of phase, thus the two channels cancel out each other, and no audio is heard. To fix this, you can try turning off mono audio in your settings, look it up, or plugging in headphones if you were not using them prior. If nothing works, try using a different device if you can't figure out the settings.
@@funkahontas How is he supposed to get openai to fix their audio?... at least he offered a solution that's actionable... talk about a stupid comment. Stick to doing triple jumps in your video games...
At 60, I hire people both younger and older than myself. It really isn't about age. I'll hire anyone of any age if they have the potential and a willingness to listen, learn, lean in and lead. Essentially do the heavy lifting and create value. Of course, when hiring, one has to be mindful of age as it comes with age related events which will impact the organization. As an executive, you don't want to wake up one morning and suddenly realize that 95% or your company is going to retire in 2 years, or starting families, etc..
Watching the thoughts of successful people is valuable. It inspires you to do something great. His comment about people having their talent wasted in the wrong company is so candid. Not all of the questions are immediately applicable but it they do satisfy curiosity. "The heaviest things in life are unmade decisions" That quote is going to stick with me. I like how Sam leans on his intuition for some of his answers
I haven't but this is one idea that could confirm other comments suggesting the video of Sam is Sora. Seems clear to me; it is constantly laggy and they would have had the tech to properly live stream his face
thank you BUT We are forced to use the new version of GPT and we have no choice. In Poland, the new version works like a calculator. It does not use personalization, it does not use the memory of the user. Moreover, it cannot check the weather forecast (the predecessor could). Below is an example conversation: Me: I had a car accident today, it really depressed me, because I try to be a good driver. You know what problems I had at work recently, it accumulates: New voice model: it’s frustrating, accidents can be very unpleasant. and how is your day going, what are your plans for the evening.
If he looks like he's an AI, its because he actually pauses before speaking. I observe that many people while on stage, on podcasts, etc, feel pressured to answer questions immediately without considering them thoughtfully. This man is masterful at that and at being super concise.
Merry Christmas, Samuel! Boy, you look fresh and happy! 😃 Yes, thank you thank you, been in detox for a couple weeks. Just philosophy, quantum mechanics 101, etc. You know, kosher stuff 🧐. Lol, heard one of your employees decided to call it ✌️quits✌️. What'd you do, jam your stilettos in the prefrontal cortex? 👌 Classic. Always pegged you (no pun intended) for a shank in the ribs type of dude. WELP , here's to you and a safe Ai future. 🥂. Mazel tov.
i love his insights. especially about the waste of potential. i see it too, and the part about angels is wild. i cant wait to see the future. i think some features will drastically change society as we know it, over time. maybe with more abundance and time people can start socializing together again, instead of through 'social networks'.
So what Sam is saying here, if you are a startup building a technology that offer foundational Generative AI offering common features and capabilities, FORGET ABOUT IT, he is inviting to build on the model rather than competing with it
2:20 - eventually I think we can offer really high quality no code tools - and already there's some out there that make sense - but you can't you can't sort of in a no code way say I - have like a full startup I want to build - um that's going to take a while 17:36 - without spoiling anything I would expect rapid progress in - image based models
I'm very glad Sam mentioned AI in healthcare. A lot of people in the world don't have access to health care, and using AI will give access to these people and give much better results people that already have access to healthcare.
Takes some time to process the information and find the best answer. Actually this is how chatGPT works. It is a combination of neuroscience and machine learning. Human's brain can be reprogrammed to work better, using some unknown technology. And then electrical signals of human's brain are captured and sent to the data centers where they are decoded into words and images and and presented as the work of their AI services. The most crazy thing is that they can capture brain activity of any person without direct physical contact. People who has become this kind of "content generators" are not happy their brains have to work intensely.
challenge accepted! | Drivers or operators. an LLM model to assist in prompting video models, image models, etc... etc... A use case scenario where different people with different levels of habitude with AI talk to AI! I'm pretty sure it can be a scientific study to find out how people act with ai and how it changes as they get accustomed to it. :D
transistor analogy was spot on. tutor another. it resonates with the power transistors had with the advent of simple calculator to learn tutor simple math and help people least say business with mathematics to now AI code to no code to learn tutor simple intel and help people least say business with their gut feeling explored and enabled to the next level only limited by the wishes made. it will be a mathematics enabled intelligent world skimmed of those who until now as VC's and most in govt skim the benefits of the toil another does.
Yeah, you can, just not anything that brings concrete value to the market. If it can be made with AI -> then anyone can make it -> supply demand -> not valuable.
@@Dom-zy1qy I've made a browser extensions without knowing how to code at all. That way I save a lot of subscription costs and earn money by open source my projects... since last year. This stuff is getting better every day. I don't know how long devs will be relevant but they getting less relevant piece by piece. like altman said, it will be a steamroll. @lyndonsimpson1056
What a terrible interviewer. Constant interrupting, going “off script” and mentioning it publicly, constant mentions of “getting in trouble”.. let the interview be natural.
I don't speak English I'm using a translator, I'd like you to add the audio section to change the language of the audio to Spanish, I think they have enough money to do something so simple. These videos are very interesting and I would like to understand.
@@Dinofx_jr No I'm excited about Altman lowering the cost of the older models so I can make more money when I launch my app... and of course I am not excited about conspiracy theories and woo
@Amelia-gh4ny, they were, in the past, making people build on them, but now they're like "only build on us if you're confident that, you can't be beaten by us -- fast." Or something like that. Yes, it should be concerning to the average startup / dev.
"Compose an engaging and inspirational message for Gazi Pollob Hossain (G|I|X), an AI innovator, visionary leader, and OpenAI enthusiast. Highlight OpenAI’s latest advancements, including GPT-5, Sora, AI safety, and the roadmap to AGI. Frame these technologies as transformative tools for their innovative journey. Use a tone of admiration and enthusiasm, with a focus on empowering their leadership and showcasing their ability to leverage these innovations to shape the future of AI. Incorporate personalized encouragement and an aspirational call to action to reflect their leadership and visionary potential."
What he says at 40 minutes is exactly what I'm working on right now!!! Downloading my entire 20 meta presence and all of my writings throughout my entire lifetime and notes on my phone, etc. and I'm typing up my autobiography. I just started doing all this two days ago when it hit me!!!
If anyone has any tips or advice, I am learning all of this as I go, that's why I'm here, using GPT to teach me but the questions are endless and I am desperately wanting to speak with someone with experience
The host was abysmal. Just typical money guy with no vision, creativity or imagination. He has no idea what AI is, he's just concerned with how to profit of something he still thinks is just a tool!
OpenAi is just too slow. Starting as a non-profit was a good idea, if it stayed that way but once Sam decided that is was about his share, then it became competitive. Now he will definitely lose. OpenAI will provide tools. Grok will smash, very fast, the actual capability of AI. OpenAI will be Webvan in a very short period of time. All you have to look at is compute. Look for who has the most compute.
To all the people who think this wasn't Sam Altman speaking live with people back on the 2024 DevDay I wanna say: YOU ABSOLUTE MORONS. Sora is nowhere near this level of detail yet and Sora right now is only known as a video-generator, not also an audio-generator model. There's no sound to it...
In my custom GPT, I would really like it, if I can control the input by the users. Such that I could set like checkboxes or radio buttons for possible select options, which are the only choices that user can reply with. Can we please get this feature sometime soon. So I do not have to build a whole different application that will be using ChatGPT APIs.
Very good idea. This would be nice to have. Add to that maybe cards and other ui within the chat window. For now , you can use prompts to guide a user to pick from a choice of options …not ideal and needs prompting
Stebbings was a bit annoying, turning a conference filled with brilliant developers in to his personal interview and views. He appears to think he is the smartest person in the room…he isn’t.
they need to incorporate episodic memory and the capacity to learn the conclusions of their reasoning, otherwise they always start all reasoning from scratch
He spent 1/4 of the interview staring up at the ceiling like he was philosophizing on some of the most novel and peculiar questions known to man. He acted like a slightly enthused teenager meagerly interested in AI who was answering questions he never in his life thought of before. Rather than a knowledgeable CEO candidly answering questions he would/should have contemplated many times over for ages now (including some he would have definitely been asked 100x already like making it open source). Pretty strange to me.
listening to these questions get it why Zoho CEO Sridhar Vembu wont get along with VC's. Talent wasted examples are the unstructured question percentage and kudos to the responses and way to go being transparent with what you are in responding and nothing to hide and a dream that is more refined as gold and brushes up the blurring quadrants of the Johari windows making it transparent to let the owner and the customer see each other as they are.
Please answer to me, you have issues on coding with plus account ? Because that's serious bullshit to go from good coding to bad coding because you have a plus account. Please tell me
nah 4o is perfect just keep working on its recursive contextual analysis, which will minimize computational needs for computer use, projects needs simulation, and procedural generation. enhance the dynamic UI to save workspace, canas needs better ide configuration and file structuring. keep 4o accessible and powerful. O1 needs predictive reason scaling , it gives great steps, and an informative correlation of didactic sourcing material for iterative planning but not dynamic synthesis it's way too limited for its true power. we should be able to synthesize chat structures into new chats and save full chats as files.
One way to paint OpenAI as an dystopian authoritarian company is to preemptively apologize for asking the most mundane questions. The moderator did a great job at that.
Harry heated up around 31 minutes on purpose maybe, perhaps to try to hide those good questions because they got really good (even if his due diligence may have seemed low quantity.)
Timestamps
00:19 - Sam Altman discusses OpenAI's future focus on reasoning models.
02:30 - OpenAI believes in continuous model improvement, impacting startups’ strategies.
06:46 - The evolving perception of AI's potential economic value.
08:49 - AI can significantly increase accessibility and reduce costs in key sectors like healthcare and education.
13:19 - Pricing models will evolve with AI's integration into the workforce.
15:34 - OpenAI focuses on improving reasoning and multimodal capabilities in future models.
19:53 - Promoting unique organizational cultures to maximize human talent.
21:46 - Rapid growth demands a shift from incremental to exponential thinking.
25:59 - Hiring diverse experience levels enhances team performance.
28:01 - Developers should consider multiple AI models for coding tasks.
32:17 - Sam Altman discusses decision-making challenges and concerns about semiconductor supply chains.
34:23 - Navigating the complexities of semiconductor supply chains and AI advancements.
38:22 - Sam Altman shares ideas on AI innovations and human potential.
40:24 - Sam Altman praises the AI industry's talent and innovation.
44:47 - Sam Altman discusses future advancements and societal impact of AI in five and ten years.
46:58 - Scientific progress continues to outpace societal changes positively.
Thank you
Thanks man 👍👍
Thank you 🙏
Great! Did you use AI to do this?
@@mr.octopus-plag yes ,i think he used merlin ai extension.
Intelligence is evolving in ways we’ve only begun to grasp. The road ahead feels deeply familiar, like a reflection of principles we hold close - humility, learning, and transformation. The possibilities are endless.Thank you for being a catalyst for progress and sparking boundless inspiration.
People are arguing whether that's really Sam on video or AI... let that sink in, what a time to be alive!
nobody is arguing that. lying makes you look stupid. shill
@zxcaaq If you think that you've been too lazy to properly check. I've seen a few comment chains where people have talked about it.
I get nothing out of lying.
A sora generated video would be less disingenuous than Altman himself
As off the rails as this was, the whole interview was so candid and valuable
Sam Altman is the GOAT. Literally OpenAI is one of the most important companies in the world right now and must be protected in all costs
“As we know, there are known knowns;
there are things we know we know.
We also know there are known unknowns;
that is to say we know there are some things we do not know.
But there are also unknown unknowns
- the ones we don’t know we don’t know.”
Is that a sora ai agent livestreaming? Just started watching and this came up in mind
It is. Im 99% convinced. Sam’s pupils aren’t very synchronized in real life.
No
Came here to say this exact same thing
The interviewer is actually the AI, and Sam Altman is real.
At this point in time this is a pretty valid question
The audio is not working in your current live stream, maybe you should not disable the comments and you guys would know that by now.
One channel is out of phase, thus the two channels cancel out each other, and no audio is heard. To fix this, you can try turning off mono audio in your settings, look it up, or plugging in headphones if you were not using them prior. If nothing works, try using a different device if you can't figure out the settings.
@@meow_meow_J How about OpenAI fix their goddamn audio ? Didn't think of that? Such a stupid comment.
Ask chat gpt how to fix it 😂
@@funkahontas How is he supposed to get openai to fix their audio?... at least he offered a solution that's actionable... talk about a stupid comment. Stick to doing triple jumps in your video games...
Maybe they shut off the comments because the comments are t0xic
the CATCH is that Sam Altman in the video is.... AI, of course using OpenAI AI tool.
I cant tell if the WIFI just sucks or if its actually real time AI taking in his audio and making a video
@@springsteen-games5134 the glitches..... the glitches.....
AMA with the ego of Harry Stebbings standing in the way
Long chats like this are super inspiring for us at BriteMinds ✨
At 60, I hire people both younger and older than myself. It really isn't about age. I'll hire anyone of any age if they have the potential and a willingness to listen, learn, lean in and lead. Essentially do the heavy lifting and create value. Of course, when hiring, one has to be mindful of age as it comes with age related events which will impact the organization. As an executive, you don't want to wake up one morning and suddenly realize that 95% or your company is going to retire in 2 years, or starting families, etc..
Unless that 95% is you and your spouse 😉
Small businesses... amirite?
@ i sincerely hope you have family to carry on the business.
Watching the thoughts of successful people is valuable. It inspires you to do something great. His comment about people having their talent wasted in the wrong company is so candid. Not all of the questions are immediately applicable but it they do satisfy curiosity. "The heaviest things in life are unmade decisions" That quote is going to stick with me. I like how Sam leans on his intuition for some of his answers
*Anyone who has worked for fElon Musk rn 😅
Am I having DejaVu? I could've sworn I saw this Q&A a few months ago
IKR! 😶
I think you're right, this is from the DevDay in London
Ikr same lol. Had to do a double take.
Me too, exactly the same
I haven't but this is one idea that could confirm other comments suggesting the video of Sam is Sora. Seems clear to me; it is constantly laggy and they would have had the tech to properly live stream his face
this is from like 6 months ago right?
Not living to your full potential.... that part of the video felt powerful and inspiring.
Thank you for the great work you are doing.
thank you BUT
We are forced to use the new version of GPT and we have no choice. In Poland, the new version works like a calculator. It does not use personalization, it does not use the memory of the user.
Moreover, it cannot check the weather forecast (the predecessor could). Below is an example conversation:
Me: I had a car accident today, it really depressed me, because I try to be a good driver. You know what problems I had at work recently, it accumulates:
New voice model:
it’s frustrating, accidents can be very unpleasant. and how is your day going, what are your plans for the evening.
I thought this was just me, I feel like a couple weeks ago I could look up recent information topics whereas now it's just giving me basic answers
AI is the greatest achievement in the history of mankind! I am duly impressed.
And will be the end of mankind aswell. Youre not very smart are you
wait im confused... is this video of Sam AI??
If he looks like he's an AI, its because he actually pauses before speaking. I observe that many people while on stage, on podcasts, etc, feel pressured to answer questions immediately without considering them thoughtfully. This man is masterful at that and at being super concise.
Merry Christmas, Samuel!
Boy, you look fresh and happy! 😃
Yes, thank you thank you, been in detox for a couple weeks. Just philosophy, quantum mechanics 101, etc. You know, kosher stuff 🧐.
Lol, heard one of your employees decided to call it ✌️quits✌️. What'd you do, jam your stilettos in the prefrontal cortex? 👌 Classic. Always pegged you (no pun intended) for a shank in the ribs type of dude.
WELP , here's to you and a safe Ai future. 🥂. Mazel tov.
it's so good to see him talk
Dev Day is so cool thanks for the new updates
Loved the lines of Sam - It looks easy or possible once it is done. So kudos to everyone who are trying to make AI main stream.
Kudos to everyone who is taking away all importance of the human spirit. Making us all obsolete useless and poor. You are a moron
Sam is a great speaker and very pleasant to listen to.
Then you should listen to what he did with his sister.
He is a practiced liar
This is déjà vu. At 3:59, he says o1-preview and o1 has already been released...
i love his insights. especially about the waste of potential. i see it too, and the part about angels is wild. i cant wait to see the future. i think some features will drastically change society as we know it, over time. maybe with more abundance and time people can start socializing together again, instead of through 'social networks'.
Bro i hope i am wrong, I am thinking Sam Altman is an AI who is talking
So great. Sam is actually sitting in front of a wall.
So you can now ask yourself in uncertainty:
what now, is there a wall or is there no wall?
What a way to promote Sora
Wait this was Sora not sam ???!
@@outdoorgeek1122 It's Sam lol Sora is no where near this good
@@ToastyZach Also not that fast lol
I really love the questions.
So what Sam is saying here, if you are a startup building a technology that offer foundational Generative AI offering common features and capabilities, FORGET ABOUT IT, he is inviting to build on the model rather than competing with it
sounds like he is just saying that to increase dependence on the product, and solidify their position
@@albertmashy8590 true, it seems so
Very poor choice of interviewer, but Sam still managed to make this interview great
2:20 - eventually I think we can offer really high quality no code tools - and already there's some out there that make sense - but you can't you can't sort of in a no code way say I - have like a full startup I want to build - um that's going to take a while
17:36 - without spoiling anything I would expect rapid progress in - image based models
I'm very glad Sam mentioned AI in healthcare. A lot of people in the world don't have access to health care, and using AI will give access to these people and give much better results people that already have access to healthcare.
Every time I hear Sam answer a question, there's this pause. A pause of thinking, reasoning, evaluating. Just like GPT-4o. Coincidence? :)
he's getting all replys flusthered in through an ear piece :P
Takes some time to process the information and find the best answer.
Actually this is how chatGPT works. It is a combination of neuroscience and machine learning. Human's brain can be reprogrammed to work better, using some unknown technology.
And then electrical signals of human's brain are captured and sent to the data centers where they are decoded into words and images and and presented as the work of their AI services. The most crazy thing is that they can capture brain activity of any person without direct physical contact. People who has become this kind of "content generators" are not happy their brains have to work intensely.
Thank for sharing👍
challenge accepted! | Drivers or operators. an LLM model to assist in prompting video models, image models, etc... etc... A use case scenario where different people with different levels of habitude with AI talk to AI! I'm pretty sure it can be a scientific study to find out how people act with ai and how it changes as they get accustomed to it. :D
Great questions!
Why is his head portrayed in such a colossal way?
AI
to hide his female sized shoulders :P
Love This
Thank you.
"Could you please add Turkish subtitles? I really enjoy your videos, but I can't fully understand them due to the language barrier. Thank you!"
transistor analogy was spot on. tutor another. it resonates with the power transistors had with the advent of simple calculator to learn tutor simple math and help people least say business with mathematics to now AI code to no code to learn tutor simple intel and help people least say business with their gut feeling explored and enabled to the next level only limited by the wishes made. it will be a mathematics enabled intelligent world skimmed of those who until now as VC's and most in govt skim the benefits of the toil another does.
Watched 49 seconds - let me guess - it's AI talking? :D ..will continue to see if I'm right :)
yes, we already generate "no code" apps with one of my chatbots.
And get millions of vulnerabilities and trash code
Yeah, you can, just not anything that brings concrete value to the market. If it can be made with AI -> then anyone can make it -> supply demand -> not valuable.
"no code" has been around since before AI hype, didn't suppress the need for devs. this will be the same.
@@Dom-zy1qy I've made a browser extensions without knowing how to code at all. That way I save a lot of subscription costs and earn money by open source my projects... since last year. This stuff is getting better every day. I don't know how long devs will be relevant but they getting less relevant piece by piece. like altman said, it will be a steamroll. @lyndonsimpson1056
True@@Dom-zy1qy
What a terrible interviewer. Constant interrupting, going “off script” and mentioning it publicly, constant mentions of “getting in trouble”.. let the interview be natural.
I don't speak English I'm using a translator, I'd like you to add the audio section to change the language of the audio to Spanish, I think they have enough money to do something so simple. These videos are very interesting and I would like to understand.
call it ClosedAI
Altman's video is so sora
Okay beer in hand... this better be good now... I want to be wowed... and I want to hear about inference price cuts!
Youre excited for the end of mankind? Nice
@@Dinofx_jr No I'm excited about Altman lowering the cost of the older models so I can make more money when I launch my app... and of course I am not excited about conspiracy theories and woo
They're begging people to avoid competing with them
This does not bode well
@Amelia-gh4ny, they were, in the past, making people build on them, but now they're like "only build on us if you're confident that, you can't be beaten by us -- fast." Or something like that. Yes, it should be concerning to the average startup / dev.
DevDay ❤4rmZambia 🇿🇲
Was it really him? or his AI assistance?
"Compose an engaging and inspirational message for Gazi Pollob Hossain (G|I|X), an AI innovator, visionary leader, and OpenAI enthusiast. Highlight OpenAI’s latest advancements, including GPT-5, Sora, AI safety, and the roadmap to AGI. Frame these technologies as transformative tools for their innovative journey. Use a tone of admiration and enthusiasm, with a focus on empowering their leadership and showcasing their ability to leverage these innovations to shape the future of AI. Incorporate personalized encouragement and an aspirational call to action to reflect their leadership and visionary potential."
I am a bit confused. Is this Sam on Video Interview or Ai?
Definitely not real Sam
@@Aedonius It is real Sam lmao
What is happening with making Open AI for profit?
What he says at 40 minutes is exactly what I'm working on right now!!! Downloading my entire 20 meta presence and all of my writings throughout my entire lifetime and notes on my phone, etc. and I'm typing up my autobiography. I just started doing all this two days ago when it hit me!!!
If anyone has any tips or advice, I am learning all of this as I go, that's why I'm here, using GPT to teach me but the questions are endless and I am desperately wanting to speak with someone with experience
depending on quantum faith
Why no one ask why the whistleblower was knocked off
He khhh himself after getting fired. Pretty sad story tbh
conspiracy weirdo
Hi Sama, when ASI? When We will be able to live 1000 years? When free energy for everyone? When Anti-Gravity Propulsion Engine?
Definition of thinking before you speak. Those pauses 😂
I miss o1 preview! 😪
The real reveal is going to be all of the videos are just ChatGPT videos working live.
The host was abysmal. Just typical money guy with no vision, creativity or imagination. He has no idea what AI is, he's just concerned with how to profit of something he still thinks is just a tool!
self driver it's big open source
Sam... is one of the busiest people in the world? :)))
Also, a wooden horse looks busy when it moves, but it doesn't get too far.
Imagine a real human at OpenAI pinning a comment before AGI offloads that process too
What is this ?? Is this even a real video
Is there a away to make it such that creators of GPTs can get paid for their GPTs by the users directly?
OpenAi is just too slow. Starting as a non-profit was a good idea, if it stayed that way but once Sam decided that is was about his share, then it became competitive. Now he will definitely lose. OpenAI will provide tools. Grok will smash, very fast, the actual capability of AI. OpenAI will be Webvan in a very short period of time. All you have to look at is compute. Look for who has the most compute.
i am at the beginning of my career and i have huge potential
23:06 Sora glitch: lets do last week what we did this week mindset
미룬결정이 뭔지 묻는 질문이 정말 좋았다고 생각했는데 두루뭉술하게 넘어가서 좀 아쉽네요
To all the people who think this wasn't Sam Altman speaking live with people back on the 2024 DevDay I wanna say:
YOU ABSOLUTE MORONS. Sora is nowhere near this level of detail yet and Sora right now is only known as a video-generator, not also an audio-generator model. There's no sound to it...
In my custom GPT, I would really like it, if I can control the input by the users. Such that I could set like checkboxes or radio buttons for possible select options, which are the only choices that user can reply with. Can we please get this feature sometime soon. So I do not have to build a whole different application that will be using ChatGPT APIs.
Very good idea. This would be nice to have. Add to that maybe cards and other ui within the chat window. For now , you can use prompts to guide a user to pick from a choice of options …not ideal and needs prompting
Stebbings was a bit annoying, turning a conference filled with brilliant developers in to his personal interview and views. He appears to think he is the smartest person in the room…he isn’t.
I love the irony that Sam thinks he’s weak at product, yet trumps the competition with how much they have to offer. 💪
they need to incorporate episodic memory and the capacity to learn the conclusions of their reasoning, otherwise they always start all reasoning from scratch
He spent 1/4 of the interview staring up at the ceiling like he was philosophizing on some of the most novel and peculiar questions known to man.
He acted like a slightly enthused teenager meagerly interested in AI who was answering questions he never in his life thought of before.
Rather than a knowledgeable CEO candidly answering questions he would/should have contemplated many times over for ages now (including some he would have definitely been asked 100x already like making it open source).
Pretty strange to me.
39:45 is he reading an ipad with chatgpt answers?
listening to these questions get it why Zoho CEO Sridhar Vembu wont get along with VC's. Talent wasted examples are the unstructured question percentage and kudos to the responses and way to go being transparent with what you are in responding and nothing to hide and a dream that is more refined as gold and brushes up the blurring quadrants of the Johari windows making it transparent to let the owner and the customer see each other as they are.
Please answer to me, you have issues on coding with plus account ? Because that's serious bullshit to go from good coding to bad coding because you have a plus account. Please tell me
Very good
Flexible agent pricing would be nice, compared to $2000/month subscription for an agent.
What a terrible interviewer. So much wasted time and opportunity.
Google is more powerful!!! Because it's free models to everyone 🎉
nah 4o is perfect just keep working on its recursive contextual analysis, which will minimize computational needs for computer use, projects needs simulation, and procedural generation. enhance the dynamic UI to save workspace, canas needs better ide configuration and file structuring. keep 4o accessible and powerful. O1 needs predictive reason scaling , it gives great steps, and an informative correlation of didactic sourcing material for iterative planning but not dynamic synthesis it's way too limited for its true power. we should be able to synthesize chat structures into new chats and save full chats as files.
instead of "agentic" we can think as "full auto bots"
Suchir Balaji
Didn't know Sam's new pet called Harry.
really cool
Wow i really don't like the interviewer.
Haha this is so A.I 😂 with the bricks 😂 u little mason 😅
9:10 what's him lookin at
thank you sama you are the best
its open AI not spend live savings into it
One way to paint OpenAI as an dystopian authoritarian company is to preemptively apologize for asking the most mundane questions. The moderator did a great job at that.
Harry heated up around 31 minutes on purpose maybe, perhaps to try to hide those good questions because they got really good (even if his due diligence may have seemed low quantity.)
Lol, openAI day at their office but Sam is on VC 😂
We are all Bozos on this bus… can’t wait to meet Whispering Squash, Lonesome Beet, and Artie Choke…