# GPT-4: A próxima evolução da inteligência artificial
## A entrevista com Sam Altman, CEO da OpenAI
### O que é o GPT-4 e como ele se diferencia do GPT-3
### ChatGPT: uma nova ferramenta de chatbot alimentada pelo GPT-4
### AGI: o futuro da inteligência artificial
## O que são leads orgânicos e como eles são diferentes de leads pagos
### A importância dos leads orgânicos para o sucesso do seu negócio
### Como diferenciar leads orgânicos de leads pagos
### Como gerar mais leads orgânicos para o seu negócio
### Estratégias eficazes para aumentar a geração de leads orgânicos
## Otimização de sites para mecanismos de busca
### A importância do SEO no sucesso do seu site
### Como otimizar seu site para mecanismos de busca
### Estratégias eficazes para melhorar o ranking do seu site
## Criação de conteúdo atraente para envolver seu público-alvo
### Como criar conteúdo de qualidade que envolve seus leitores
### Estratégias eficazes para criar conteúdo engajador
## Utilização das redes sociais para promover conteúdo e expandir sua audiência
### A importância das redes sociais no marketing digital
### Como utilizar as redes sociais para promover seu conteúdo
### Estratégias eficazes para expandir sua audiência nas redes sociais
## Construção de uma lista de e-mails efetiva para marketing
### A importância da construção de uma lista de e-mails
### Como criar uma lista de e-mails efetiva
### Estratégias eficazes para aumentar a lista de e-mails
## Conclusão
### O GPT-4 é uma evolução revolucionária da inteligência artificial
### Leads orgânicos são importantes para o sucesso do seu negócio
### Otimização de sites, criação de conteúdo, utilização das redes sociais e construção de listas de e-mails são estratégias eficazes
## Perguntas frequentes sobre GPT-4, leads orgânicos e marketing digital
1. Como funciona o GPT-4?
2. Qual a diferença entre leads orgânicos e leads pagos?
3. Como gerar mais leads orgânicos para meu negócio?
4. Como otimizar meu site para mecanismos de busca?
5. Qual a importância das redes sociais para o marketing digital?

br>¿Cuándo saldrá GPT-4? ¿Es ChatGPT el fin de Google? ¿Cómo será el futuro de las AGIs? Todas estas preguntas se encuentran en la siguiente entrevista a Sam Altman, realizada el pasado mes de Enero por StriclyVC. Por el interés en su contenido, hoy os traigo un resumen subtitulado al castellano para que vosotros también podáis escuchar a una de las personas más relevantes del mundo tecnológico, el CEO de OpenAI.

► Entrevista a Sam Altman, StriclyVC (PARTE 1 – Inversiones)

► Entrevista a Sam Altman, StriclyVC (PARTE 2 – OpenAI e IA)

► Helion y Fusión Nuclear, Real Engineering

► Los problemas con Helion Energy

📹 EDICIÓN: Carlos Santana y Diego Gonzalez (Diocho)

— ¡MÁS DOTCSV! —-

📣 NotCSV – ¡Canal Secundario!

💸 Patreon :
👓 Facebook :
👾 Twitch!!! :
🐥 Twitter :
📸 Instagram :

— ¡MÁS CIENCIA! —

🔬 Este canal forma parte de la red de divulgación de SCENIO. Si quieres conocer otros fantásticos proyectos de divulgación entra aquí:

Este vídeo foi indexado através do Youtube link da fonte
gpt 4 ,

dotcsv,machine learning,inteligencia artificial,deep learning,informática,programación,dot csv,csv,carlos santana vega,carlos santana,dot,openai,sam altman,open ai,entrevista a sam altman ,

https://www.youtubepp.com/watch?v=LbWjVNjlpjA ,

I yeah it’s just taken up a lot of my time but it’s super great and I think we’re we’re doing a lot of stuff we’re really proud of um I don’t think people are interested in open AI because of my involvement I think open AI has managed to pull

Together the the most Talent dense researchers and engineers in the field of AI who have done just like incredible work and I think what people are interested in is like open AI from a cold start a few years ago has managed to do this thing that I think is going

To be incredibly important to the many decades at least of society and how we all live our lives and what we do and what’s possible and I think it’s going to be tremendously good questions why do you think that chat GPT and Dolly so surprised people I genuinely don’t

Know I’ve reflected on it a lot we had the we had a model in the we had the model for cha gbt in the API for like I don’t know 10 months or something before we made Chachi PT and I sort of thought someone was going to just build it or whatever

And that you know enough people had played around with it definitely if you make a really good user experience on top of something like one thing that I very deeply believed was the way people wanted to interact with these models was via dialogue and I you know we kept

Telling people this we kept trying to get these people to build it and people wouldn’t quite do it so we finally said all right we’re just going to do it yeah I think the pieces were there for a while um one of the reasons I think Dolly

Surprised people is if you asked you know five or seven years ago that that kind of Ironclad wisdom on AI first it comes for physical labor truck driving work in the factory working Factory then truck driving then this sort of um less demanding cognitive labor than the really demanding cognitive labor like

Computer programming and then and then very last of all or maybe never because maybe it’s like some deep human Special Sauce was creativity and of course we can look now and say it really looks like it’s going to go exactly the direction but I think that is not super

Intuitive and so I can see why Dolly surprised people yesterday is um I would have expected maybe like one order of magnitude less of everything like one order and two less of um hype one order magnitude less of users but I genuinely felt somewhat confused about why Chachi BT did uh you

Know one of the things we really believe is that the most responsible way to put this out in society is very gradually and to get people institutions policy makers get them familiar with it thinking about the implications feeling the technology getting a sense for what

It can do and can’t do rather than drop a super powerful AGI in the world all at once and so we put GPT 3 out almost three years ago and then we put it into an API like you know kind of I think it was maybe like June of two like so two

And a half years ago the the incremental update from that to Chachi BT I felt like should have been predictable and I want to like do more introspection on why I was sort of miscalibrated on that we do a we have like an internal process

Where we kind of try to break things and and study impacts we we use external Auditors we have external red teamers we we work with other labs and have safety organizations look at stuff chat gbt is going to cause or is causing there’s I think a big one going now about the

Impact of this on education academic Integrity all of that but starting these now where the stakes are still relatively low rather than just put out what the whole industry will have in a few years with no time for society to update I think would be bad

Kovid did show us For Better or For Worse or at least me that Society can update to like massive changes sort of faster than I would have thought in many ways but I still think like given the magnitude of the economic impact we expect here more gradual is better and

So putting out a very weak and imperfect system like Chao gbt and then making it a little better this year a little better later this year a little better next year that seems much better than the alternative yes I get it um I get why Educators feel the way they

Feel about this and and probably like this is just a preview of what we’re going to see in a lot of other areas um I think this is just the new we’re going to try to you know do some things in the short term um and there may be ways we can help

Teachers be like a little bit more likely to detect output or anyone output of like a gpt-like system but honestly a determined person is going to get around them and I don’t think it’ll be something Society can or should rely on long term we’re just in a new world now

Like generated text is something we all need to adapt to and that’s fine we adapted to you know calculators and changed what we tested for in math classes I imagine uh this is a more extreme version of that no doubt but also the benefits of it are are more Extreme as well

Um you know we hear about we hear from teachers who are understandably very nervous about the impact of this on homework we also hear a lot from teachers who are like wow this is like an unbelievable personal tutor for each kid and I think that I have used it to

Learn things myself uh and found it like much more compelling than other ways I’ve learned things in the past like I would much rather have Chachi BT teaching me about something than go read a textbook uh so you know it’s like an evolving world and we’ll

All adapt and I think be better off for it and we won’t want to go back yes Professor is you know we will we will experiment with this other people will too I think it is important for the transition um but I I would caution policy National policy individual schools or

Whatever from relying on this um because I don’t like fundamentally I think it’s impossible to make it perfect you’ll think you know people will figure out how much of the text they have to change there will be other things that modify the outputted text um

I think it’s good to pursue and we will um but I think what’s important to realize is like the playing field has shifted and that’s fine there’s good and bad and we just figure out like rather than try to go back we figure out the way forward

It’ll come out at some point when we are like confident that we can do it safely and responsibly I think in general we are going to release technology much more slowly than people would like we’re going to sit on it for much longer than people would like and eventually people

Will be like happy with our approach to this but at the time I realized like people want the shiny toy and it’s frustrating I totally get that I saw a visual and I don’t know if it was accurate but it showed GPT 3.5 versus I guess what gpt4 is expected and I saw

That thing on Twitter did you was that was it no okay that was a little bit scary um the gbt4 rumor mill is like a ridiculous thing I don’t know where it all comes from I don’t know why people don’t have like better things to

Speculate on I get a little bit of it like it’s sort of fun but that it’s been going for like six months at this volume people are begging to be disappointed and they will be like it’s you know people are gonna like the hype is just

Like we don’t have an actual AGI and I think that’s sort of what is expected of us and you know yeah we’re going to disappoint those people video is that coming it will come I wouldn’t want to make a confident prediction about when um obviously like

People are interested in it uh we’ll try to do it other people will try to do it um it could be like pretty soon it’s a it’s a legitimate research project so it could be pretty soon it could take a while okay it is the future I would like to see is where

Access to AI is super democratized where there are several agis in the world that can kind of like help allow for multiple viewpoints and not have anyone get too powerful and that the cost of intelligence and energy because it gets commoditized Trends down and down and down and the massive Surplus there

Access to the systems eventually governance of the systems benefits all of us so yeah I sort of hope that happens I think competition is good at least you know until we get to AGI I like deeply believe in in capitalism and competition to offer the best service at

The lowest price what I think is going to have to happen is society will have to agree and like set some laws on what an AGI can never do or what one of these systems should never do and one of the cool things about the the path of the

Technology tree that we’re on which is very different like before we came along and it was sort of deepmind having these games that were like you know having agents play each other and try to deceive each other and kill each other and all that which I think could have

Gone in a bad direction we know these language models that can uh understand language and so we can say Hey you know model here’s what we’d like you to do here are the values we’d like you to align to and we don’t have it working perfectly yet but it works a

Little and it’ll get better and better and the world can say all right here are the rules here’s here’s the very broad bounds of very broad like absolute rules of a system but within that people should be allowed very different things that they want

Their AI to do and so if you want the super like you know never offend say for work model you should get that and if you want an edgier one that you know is sort of like creative and exploratory but says some stuff you like might not be comfortable

With or some people might not be comfortable with you should get that um and I think there will be many systems in the world that have different settings of the values that they enforce and really what I think and this will take longer is that

You as a user should be able to write up a few pages of here’s what I want here are my values here’s how I want the AI to behave and it reads it and thinks about it and acts exactly um how you want because it’s like should

Be your AI uh and you know it should be there to serve you and do the things you believe in so that to me is much better than one system where like one tech company says here are the rules I think Society should regulate what the kind of

The wide bounds are but then I think individual users should have a huge amount of Liberty to decide how they want their experience and their interaction to go so I think it is like a combination of Society you know like we have there are a few asterisks on the Free Speech rules

Um and Society has decided like free speech not quite absolute I think Society will decide also decide like language model is not quite absolute but there are a lot of there’s a lot of speech that is legal that you find distasteful that I find distasteful that he finds distasteful and we all probably

Have somewhat different definitions of that and I think it is very important that that is left to the responsibility of individual users and groups not not one company and that the government they are like govern and not dictate all of the rules I mean I I think the best case is like

So unbelievably good that it’s like hard to I I think it’s like hard for me to even imagine like I can sort of I can sort of think about what it’s like when we make more progress of discovering new new knowledge with these systems then

Humanity has done so far but like in a year instead of seventy thousand um I can sort of imagine what it’s like when we kind of like launch probes out to the whole universe and find out really you know everything going on out there I can sort of imagine what it’s

Like when we have just like unbelievable abundance and systems that can sort of you know help us resolve Deadlocks and improve all aspects of reality and uh kind of like let us all live our best lives but I can’t quite like I think the the the good case is just so

Unbelievably good that you sound like a really crazy person to start talking about it um and the bad case and I think this is like important to say is like lights out for all of us I’m more worried about like an accidental misuse case in the

Short term where you know someone gets a super powerful like it’s not like the AI wakes up and decides to be evil and I think all of the sort of traditional AI safety thinkers reveal a lot about more than about themselves than they mean to when they

Talk about what they think the AGI is going to be like but but I can see The Accidental misuse case clearly and that’s that’s super bad um so I think like uh yeah I think it’s like impossible to overstate the importance of AI safety and Alignment work I would like to see

Much much more happening but I think it’s more subtle than most people think the closer we get the harder time I have answering because I think that it’s going to be much blurrier and much more of a gradual transition than than people think if you if you imagine like a two

By two Matrix of sort of short timelines until the AGI takeoff era begins and long timelines until it begins and then a slow take off or a fast take off the world I think we’re heading to and the safest world the one I most hope for is

The short timeline slow takeoff but I think people are going to have hugely different opinions about when and there you kind of like declare Victory on the AGI thing is opinion I think Ubi is good and important but uh very far from sufficient I think it is like a little

Part of the solution I think it’s great like I think we should as AGI participates more and more in the economy I think we should distribute wealth and and resources much more than than we have um and that’ll be important over time but I don’t I don’t think that’s gonna like

Solve the problem I don’t think that’s going to give people meaning I don’t think it means people are going to like entirely stop trying to create and do new things and whatever else so I sort of would consider it like enabling technology but not like a plan for

Society can I ask selfishly so if Ubi is only part of the solution and I’ve got teenagers and we all have jobs what should we be preparing for you know as I said my my son’s teacher was trying to prepare them but of course you would maybe be better positioned to have some

Ideas on this resilience adaptability ability to like learn new things quickly creativity although it’ll be aided creativity like in some sense before Google came along um there was like a bunch of things that we learned uh like memorizing facts was really important and that changed and now I think learning will change again

And we’ll probably adapt faster than we think like again in general we are very much here to build AGI um and products and services are tactics in service of that Partnerships too um but important ones and we like we really want to be useful to people and I

Think if we just build this in a lab and don’t figure out how to get out into the world uh that’s that’s like somehow we’re really falling short there is well I wondered what you made of the fact that Google has said to its employees we just it’s too

Imperfect it could harm our reputation we’re not ready I hope I hope when they launch something anyway you really hold them to that comment um I’ll just leave it there yes yeah I think whenever someone like talks about a technology being the end of some other giant company it’s usually wrong

Like I think people they get to make a counter move here and they’re like pretty smart pretty competent but I do think it me there is a change for search that will probably come at some point but not as dramatically as people think in the short term would like my guess is

That people are going to be using Google the same way people are using Google Now for quite some time and also Google for whatever this whole like code red thing is is probably not going to change that dramatic would be my guess is it’s great they’re the only tech company

Out there that I think we I’d be excited to partner with this deeply um I think Satya is an amazing CEO but more than that human being and understands the stakes of what AGI means and why we need to have all the weirdness we do in our structure and our agreement with them

And so I really feel like it’s a very values aligned company and there’s some things they’re very good at like building very large supercomputers uh and the infrastructure we operate on and putting the technology into products there’s things we’re very good at like doing research um and it’s been a great partnership can

You comment on whether the reports are accurate that it’s going to be in bing office or maybe it’s already in those things you are a very experienced and professional reporter you know I can’t comment on that I know you know I can’t comment on that you

Know I know you know you can’t comment on that in the spirit of shortness of life and our precious time here why do you ask Sam I’m genuinely curious well I’m like if you ask a question you know I’m not going to answer well I thought you might

Answer that one no okay I know there’s some things you don’t answer but I gotta try another company’s product plans I’m definitely not going to touch it is I I try to just do things that I’m interested in at this point and so I’ve tried to like hold myself to the stuff

That I really love which tends to be like the hard tech years of r d Capital intensive or like sort of like risky art like risky research and then but if it works it really works helion is like more than an investment I mean that’s that’s the other thing besides open AI I

Spend a lot of a lot of time on and uh just super excited about what’s gonna happen there not not sort of getting Fusion to work in a lab although that that is cool too um but Building A system that will work at a super low cost so if you look at

The previous energy transitions if you can get the cost of a new form of energy down it can like take over everything else in a couple of decades just phenomenally fast and then also a system where we can create enough energy and enough reliable energy both in terms of

The machines not breaking and also not having like the intermittents or the need for storage of of solar wind or something like that if we can create enough for Earth in like 10 years and I think that’s actually the hardest challenge that helion faces um as as we sketch out what it takes

Operationally to do that to replace all the current generative capacity on Earth with fusion and to do it really fast and to like think about what it really means to build a factory that’s capable of like putting out two of these machines a day for a decade that’s really hard but

Also a super fun problem so I think there will be I’m very happy there’s a fusion race I think that’s great I’m also very happy solar and batteries are getting so cheap but I think what will matter is who can deliver energy uh the cheapest and enough of it I think

Probably will work either um but to what I was just saying earlier I think it will be commercially irrelevant they also think it’ll be commercially irrelevant um the thing that is so exciting to me about helion is that that it’s a simple machine that is an affordable cost and a

Reasonable size there’s a bunch of different elements of it than like the giant tokamax but one that is very cool is what what comes out of the reaction is charged particles not heat almost all other sort of like you know like a coal plant or a natural gas plant or whatever

Makes heat drives a steam turbine that’s what it does helion makes charged particles which push back on the magnet and drive an electrical current down a wire there’s no heat cycle at all and so it can be a much simpler much more efficient system and that I think is

Like Mist out of the whole discussion on Fusion but really great it also means we don’t have to deal with like much nuclear material we don’t ever have like dangerous waste or even a dangerous system you could like touch it pretty shortly after it turns off has it proven it’s thesis

We will have more to share there shortly I think by like 2028 pending you know good fortune with Regulators we could be plugging them into the grid I think we’ll do it you know a really great demo well before that like hopefully pretty soon is foreign

,00:00 I yeah it’s just taken up a lot of my
00:01 time but it’s super great and I think
00:03 we’re we’re doing a lot of stuff we’re
00:05 really proud of
00:08 um
00:35 I don’t think people are interested in
00:38 open AI because of my involvement I
00:40 think open AI has managed to pull
00:41 together the the most Talent dense
00:43 researchers and engineers in the field
00:46 of AI who have done just like incredible
00:49 work and I think what people are
00:50 interested in is like open AI from a
00:52 cold start a few years ago has managed
00:55 to do this thing that I think is going
00:57 to be incredibly important to the
01:01 many decades at least of society and how
01:04 we all live our lives and what we do and
01:06 what’s possible and I think it’s going
01:07 to be tremendously good questions
01:48 why do you think that chat GPT and Dolly
01:52 so surprised people I genuinely don’t
01:55 know I’ve reflected on it a lot
01:58 we had the we had a model in the we had
02:00 the model for
02:02 cha gbt in the API for like I don’t know
02:04 10 months or something before we made
02:06 Chachi PT and I sort of thought someone
02:08 was going to just build it or whatever
02:10 and that you know enough people had
02:11 played around with it definitely if you
02:13 make a really good user experience on
02:16 top of something like one thing that I
02:18 very deeply believed was the way people
02:21 wanted to interact with these models was
02:23 via dialogue and I you know we kept
02:26 telling people this we kept trying to
02:27 get these people to build it and people
02:28 wouldn’t quite do it so we finally said
02:29 all right we’re just going to do it yeah
02:30 I think the pieces were there for a
02:32 while
02:33 um one of the reasons I think Dolly
02:34 surprised people is if you asked you
02:36 know five or seven years ago that that
02:38 kind of Ironclad wisdom on AI first it
02:40 comes for physical labor truck driving
02:42 work in the factory working Factory then
02:44 truck driving then this sort of um less
02:47 demanding cognitive labor than the
02:49 really demanding cognitive labor like
02:50 computer programming and then and then
02:52 very last of all or maybe never because
02:54 maybe it’s like some deep human Special
02:57 Sauce was creativity and of course we
02:59 can look now and say it really looks
03:01 like it’s going to go exactly the
03:02 direction but I think that is not super
03:05 intuitive and so I can see why Dolly
03:07 surprised people yesterday is
04:15 um I would have expected maybe like one
04:17 order of magnitude less of everything
04:19 like one order and two less of
04:21 um hype one order magnitude less of
04:23 users but I genuinely felt somewhat
04:26 confused about why Chachi BT did uh you
04:29 know one of the things we really believe
04:31 is that the most responsible way to put
04:33 this out in society is very gradually
04:34 and to get people institutions policy
04:39 makers get them familiar with it
04:40 thinking about the implications feeling
04:42 the technology getting a sense for what
04:43 it can do and can’t do rather than drop
04:45 a super powerful AGI in the world all at
04:47 once and so we put GPT 3 out almost
04:51 three years ago and then we put it into
04:53 an API like you know kind of I think it
04:55 was maybe like June of two like so two
04:57 and a half years ago the the incremental
04:59 update from that to Chachi BT I felt
05:01 like should have been predictable and I
05:03 want to like do more introspection on
05:05 why I was sort of miscalibrated on that
05:06 we do a we have like an internal process
05:09 where we kind of try to break things and
05:10 and study impacts we we use external
05:13 Auditors we have external red teamers we
05:15 we work with other labs and have safety
05:17 organizations look at stuff chat gbt is
05:19 going to cause or is causing there’s I
05:22 think a big one going now about the
05:24 impact of this on education academic
05:27 Integrity all of that but
05:29 starting these now where the stakes are
05:33 still relatively low rather than just
05:36 put out what the whole industry will
05:37 have in a few years with no time for
05:39 society to update I think would be bad
05:41 kovid did show us For Better or For
05:44 Worse or at least me that Society can
05:46 update to like massive changes sort of
05:49 faster than I would have thought in many
05:51 ways but I still think like given the
05:54 magnitude of the economic impact we
05:55 expect here more gradual is better and
05:58 so putting out a very weak and imperfect
06:00 system like Chao gbt and then making it
06:02 a little better this year a little
06:03 better later this year a little better
06:05 next year that seems much better than
06:07 the alternative yes
07:16 I get it
07:19 um I get why Educators feel the way they
07:22 feel about this and and probably like
07:24 this is just a preview of what we’re
07:26 going to see in a lot of other areas
07:28 um I think this is just the new we’re
07:30 going to try to you know do some things
07:33 in the short term
07:35 um
07:35 and there may be ways we can help
07:37 teachers be like a little bit more
07:39 likely to detect output or anyone output
07:42 of like a gpt-like system but honestly a
07:45 determined person is going to get around
07:46 them and I don’t think it’ll be
07:48 something Society can or should rely on
07:50 long term we’re just in a new world now
07:52 like generated text is something we all
07:55 need to adapt to
07:57 and that’s fine we adapted to you know
08:01 calculators and changed what we tested
08:03 for in math classes I imagine uh this is
08:05 a more extreme version of that no doubt
08:07 but also the benefits of it are are more
08:10 Extreme as well
08:12 um you know we hear about we hear from
08:15 teachers who are understandably very
08:16 nervous about the impact of this on
08:18 homework we also hear a lot from
08:20 teachers who are like wow this is like
08:21 an unbelievable personal tutor for each
08:23 kid and I think that I have used it to
08:26 learn things myself uh
08:28 and found it like much more compelling
08:30 than other ways I’ve learned things in
08:31 the past like I would much rather have
08:34 Chachi BT teaching me about something
08:35 than go read a textbook uh so you know
08:39 it’s like an evolving world and we’ll
08:42 all adapt and I think be better off for
08:44 it and we won’t want to go back
08:48 yes
08:58 Professor is
09:30 you know we will we will experiment with
09:32 this other people will too I think it is
09:34 important for the transition
09:36 um but I I would caution
09:39 policy
09:40 National policy individual schools or
09:42 whatever from relying on this
09:45 um because I don’t like
09:47 fundamentally I think it’s impossible to
09:49 make it perfect you’ll think you know
09:51 people will figure out how much of the
09:53 text they have to change there will be
09:54 other things that modify the outputted
09:56 text
09:57 um
09:58 I think it’s good to pursue and we will
10:03 um
10:03 but I think what’s important to realize
10:05 is like the playing field has shifted
10:07 and that’s fine there’s good and bad and
10:10 we just figure out like rather than try
10:12 to go back we figure out the way forward
10:32 it’ll come out at some point when we are
10:34 like confident that we can do it safely
10:37 and responsibly I think in general we
10:39 are going to release technology much
10:42 more slowly than people would like we’re
10:44 going to sit on it for much longer than
10:45 people would like and eventually people
10:47 will be like happy with our approach to
10:49 this but at the time I realized like
10:51 people want the shiny toy and it’s
10:52 frustrating I totally get that I saw a
10:54 visual and I don’t know if it was
10:55 accurate but it showed GPT 3.5 versus I
10:59 guess what gpt4 is expected and I saw
11:01 that thing on Twitter did you was that
11:03 was it no okay that was a little bit
11:05 scary
11:06 um the gbt4 rumor mill is like
11:09 a ridiculous thing I don’t know where it
11:12 all comes from I don’t know why people
11:13 don’t have like better things to
11:14 speculate on I get a little bit of it
11:16 like it’s sort of fun but that it’s been
11:18 going for like six months at this volume
11:19 people are begging to be disappointed
11:21 and they will be like it’s you know
11:23 people are gonna like the hype is just
11:25 like we don’t have an actual AGI and I
11:27 think that’s sort of what is expected of
11:29 us and you know yeah we’re going to
11:30 disappoint those people video
11:33 is that coming
11:34 it will come I wouldn’t want to make a
11:36 confident prediction about when
11:38 um obviously like
11:41 people are interested in it uh we’ll try
11:44 to do it other people will try to do it
11:47 um it could be like pretty soon it’s a
11:49 it’s a legitimate research project so it
11:50 could be pretty soon it could take a
11:52 while okay it is
12:07 the future I would like to see is where
12:11 access to AI is super democratized where
12:15 there are several agis in the world that
12:18 can kind of like help allow for multiple
12:21 viewpoints and not have anyone get too
12:24 powerful and that the cost of
12:28 intelligence and energy because it gets
12:31 commoditized Trends down and down and
12:33 down and the massive Surplus there
12:35 access to the systems eventually
12:38 governance of the systems benefits all
12:40 of us so yeah I sort of hope that
12:43 happens I think competition is good at
12:45 least you know until we get to AGI I
12:47 like deeply believe in in capitalism and
12:49 competition to offer the best service at
12:51 the lowest price what I think is going
12:53 to have to happen is society will have
12:56 to agree and like set some laws on what
13:00 an AGI can never do or what one of these
13:02 systems should never do and one of the
13:04 cool things about the the path of the
13:06 technology tree that we’re on which is
13:08 very different like before we came along
13:10 and it was sort of deepmind having these
13:11 games that were like you know having
13:13 agents play each other and try to
13:15 deceive each other and kill each other
13:16 and all that which I think could have
13:17 gone in a bad direction we know these
13:19 language models that can uh
13:22 understand language and so we can say
13:24 Hey you know model here’s what we’d like
13:27 you to do here are the values we’d like
13:28 you to align to and we don’t have it
13:31 working perfectly yet but it works a
13:32 little and it’ll get better and better
13:33 and the world can say all right here are
13:36 the rules here’s here’s the very broad
13:38 bounds of very broad like
13:42 absolute rules of a system
13:44 but within that people should be allowed
13:47 very different things that they want
13:48 their AI to do and so if you want the
13:51 super like
13:53 you know never offend say for work model
13:55 you should get that and if you want an
13:57 edgier one that you know is sort of like
14:00 creative and exploratory but says some
14:02 stuff you like might not be comfortable
14:05 with or some people might not be
14:06 comfortable with you should get that
14:08 um and I think there will be many
14:09 systems in the world that have different
14:11 settings of the values that they enforce
14:14 and really what I think and this will
14:16 take longer is that
14:18 you as a user should be able to write up
14:20 a few pages of here’s what I want here
14:24 are my values here’s how I want the AI
14:25 to behave and it reads it and thinks
14:27 about it and acts exactly
14:29 um how you want because it’s like should
14:31 be your AI uh and you know it should be
14:35 there to serve you and do the things you
14:36 believe in so that to me is much better
14:39 than one system where like one tech
14:41 company says here are the rules I think
14:43 Society should regulate what the kind of
14:47 the wide bounds are but then I think
14:50 individual users should have a huge
14:52 amount of Liberty to decide how they
14:56 want their experience and their
14:58 interaction to go so I think it is like
15:00 a combination of
15:03 Society you know like we have there are
15:05 a few asterisks on the Free Speech rules
15:09 um and Society has decided like free
15:11 speech not quite absolute I think
15:12 Society will decide also decide like
15:14 language model is not quite absolute but
15:17 there are a lot of there’s a lot of
15:19 speech that is legal that you find
15:21 distasteful that I find distasteful that
15:22 he finds distasteful and we all probably
15:24 have somewhat different definitions of
15:26 that and I think it is very important
15:29 that that is left
15:31 to the responsibility of individual
15:34 users and groups not not one company and
15:36 that the government they are like govern
15:38 and not dictate all of the rules
15:47 I mean I I think the best case is like
15:50 so unbelievably good that it’s like hard
15:53 to
15:54 I I think it’s like hard for me to even
15:56 imagine like I can sort of I can sort of
15:59 think about what it’s like when we make
16:01 more progress of discovering new new
16:04 knowledge with these systems then
16:06 Humanity has done so far but like in a
16:09 year instead of seventy thousand
16:11 um I can sort of imagine what it’s like
16:12 when we kind of like launch probes out
16:15 to the whole universe and find out
16:18 really you know everything going on out
16:19 there I can sort of imagine what it’s
16:20 like when we have just like unbelievable
16:23 abundance and systems that can sort of
16:26 you know help us resolve Deadlocks and
16:29 improve all aspects of reality and uh
16:33 kind of like let us all live our best
16:34 lives but I can’t quite like I think the
16:37 the the good case is just so
16:39 unbelievably good that you sound like a
16:42 really crazy person to start talking
16:44 about it
16:45 um
16:47 and the bad case and I think this is
16:48 like important to say is like lights out
16:50 for all of us I’m more worried about
16:54 like an accidental misuse case in the
16:58 short term where you know someone gets a
17:00 super powerful like it’s not like the AI
17:02 wakes up and decides to be evil and I
17:04 think all of the sort of traditional AI
17:06 safety thinkers
17:07 reveal a lot about more than about
17:10 themselves than they mean to when they
17:11 talk about what they think the AGI is
17:13 going to be like but but I can see The
17:15 Accidental misuse case clearly and
17:18 that’s that’s super bad
17:20 um so I think like
17:22 uh yeah I think it’s like impossible to
17:25 overstate the importance of AI safety
17:27 and Alignment work I would like to see
17:30 much much more happening but I think
17:32 it’s more subtle than most people think
17:39 the closer we get the harder time I have
17:42 answering because I think that it’s
17:44 going to be much blurrier and much more
17:47 of a gradual transition than than people
17:49 think if you if you imagine like a two
17:51 by two Matrix of sort of short timelines
17:55 until the AGI takeoff era begins and
17:59 long timelines until it begins and then
18:01 a slow take off or a fast take off the
18:03 world I think we’re heading to and the
18:04 safest world the one I most hope for is
18:06 the short timeline slow takeoff but I
18:10 think people are going to have hugely
18:13 different opinions about when and there
18:14 you kind of like declare Victory on the
18:16 AGI thing is
18:34 opinion I think Ubi is
18:36 good and important but uh very far from
18:40 sufficient I think it is like a little
18:42 part of the solution I think it’s great
18:44 like I think we should as
18:46 AGI participates more and more in the
18:49 economy I think we should distribute
18:51 wealth and and resources much more than
18:54 than we have
18:55 um
18:56 and that’ll be important over time but I
18:59 don’t I don’t think that’s gonna like
19:02 solve the problem I don’t think that’s
19:04 going to give people meaning I don’t
19:05 think it means people are going to like
19:06 entirely stop trying to create and do
19:07 new things and whatever else so I sort
19:09 of would consider it like enabling
19:11 technology but not like a plan for
19:13 society can I ask selfishly so if Ubi is
19:16 only part of the solution and I’ve got
19:18 teenagers and we all have jobs what
19:21 should we be preparing for you know as I
19:22 said my my son’s teacher was trying to
19:25 prepare them but of course you would
19:27 maybe be better positioned to have some
19:30 ideas on this resilience adaptability
19:32 ability to like learn new things quickly
19:35 creativity although it’ll be aided
19:37 creativity like in some sense before
19:40 Google came along
19:41 um there was like a bunch of things that
19:42 we learned uh like memorizing facts was
19:45 really important and that changed and
19:47 now I think learning will change again
19:49 and
19:51 we’ll probably adapt faster than we
19:53 think like again in general we are very
19:55 much here to build AGI
19:57 um and products and services are tactics
20:00 in service of that Partnerships too
20:03 um but important ones and we like we
20:06 really want to be useful to people and I
20:08 think if we just build this in a lab and
20:09 don’t figure out how to get out into the
20:11 world uh that’s that’s like somehow
20:14 we’re really falling short there is
20:35 well I wondered what you made of the
20:37 fact that Google has said
20:39 to its employees we just it’s too
20:41 imperfect it could harm our reputation
20:43 we’re not ready I hope I hope when they
20:46 launch something anyway
20:48 you really hold them to that comment
20:51 um
20:52 I’ll just leave it there
20:54 yes
21:34 yeah I think whenever someone like talks
21:37 about a technology being the end of some
21:38 other giant company it’s usually wrong
21:40 like I think people they get to make a
21:43 counter move here and they’re like
21:44 pretty smart pretty competent but I do
21:47 think it me there is a change for search
21:50 that will probably come at some point
21:52 but not as dramatically as people think
21:55 in the short term would like my guess is
21:56 that people are going to be using Google
21:58 the same way people are using Google Now
22:00 for quite some time and also Google for
22:02 whatever this whole like code red thing
22:03 is is probably not going to change that
22:05 dramatic would be my guess is
22:23 it’s great they’re the only tech company
22:25 out there that I think we I’d be excited
22:28 to partner with this deeply um I think
22:30 Satya is an amazing CEO but more than
22:32 that human being and understands the
22:34 stakes of what AGI means and why we need
22:37 to have all the weirdness we do in our
22:39 structure and our agreement with them
22:41 and so I really feel like it’s a very
22:43 values aligned company and there’s some
22:45 things they’re very good at like
22:46 building very large supercomputers uh
22:48 and the infrastructure we operate on and
22:49 putting the technology into products
22:51 there’s things we’re very good at like
22:52 doing research
22:54 um and it’s been a great partnership can
22:56 you comment on whether the reports are
22:57 accurate that it’s going to be in bing
22:59 office or maybe it’s already in those
23:01 things
23:07 you are a very experienced and
23:08 professional reporter
23:10 you know I can’t comment on that I know
23:12 you know I can’t comment on that you
23:14 know I know you know you can’t comment
23:15 on that in the spirit of shortness of
23:18 life and our precious time here why do
23:20 you ask
23:23 Sam I’m genuinely curious well I’m like
23:26 if you ask a question you know I’m not
23:28 going to answer well I thought you might
23:29 answer that one no okay I know there’s
23:31 some things you don’t answer but I gotta
23:32 try another company’s product plans I’m
23:34 definitely not going to touch it
23:41 is
24:14 I I try to just do things that I’m
24:17 interested in at this point and so I’ve
24:19 tried to like hold myself to the stuff
24:21 that I really love which tends to be
24:22 like the hard tech years of r d Capital
24:26 intensive or like sort of like risky art
24:30 like risky research and then but if it
24:32 works it really works helion is like
24:34 more than an investment I mean that’s
24:36 that’s the other thing besides open AI I
24:37 spend a lot of a lot of time on and uh
24:40 just super excited about what’s gonna
24:41 happen there not not sort of getting
24:43 Fusion to work in a lab although that
24:45 that is cool too
24:46 um but Building A system that will work
24:49 at a super low cost so if you look at
24:52 the previous energy transitions if you
24:54 can get the cost of a new form of energy
24:56 down it can like take over everything
24:57 else in a couple of decades just
24:59 phenomenally fast and then also a system
25:02 where we can create enough energy and
25:04 enough reliable energy both in terms of
25:05 the machines not breaking and also not
25:07 having like the intermittents or the
25:09 need for storage of of solar wind or
25:11 something like that if we can create
25:12 enough for Earth in like 10 years and I
25:15 think that’s actually the hardest
25:16 challenge that helion faces
25:18 um as as we sketch out what it takes
25:21 operationally to do that to replace all
25:23 the current generative capacity on Earth
25:25 with fusion and to do it really fast and
25:28 to like think about what it really means
25:30 to build a factory that’s capable of
25:31 like putting out two of these machines a
25:33 day for a decade that’s really hard but
25:36 also a super fun problem so I think
25:38 there will be I’m very happy there’s a
25:40 fusion race I think that’s great I’m
25:43 also very happy solar and batteries are
25:44 getting so cheap but I think what will
25:46 matter is who can deliver energy uh the
25:49 cheapest and enough of it I think
25:51 probably will work either
25:53 um but to what I was just saying earlier
25:55 I think it will be commercially
25:56 irrelevant they also think it’ll be
25:58 commercially irrelevant
26:00 um the thing that is so exciting to me
26:01 about helion is that that
26:05 it’s a simple machine
26:07 that is an affordable cost and a
26:10 reasonable size there’s a bunch of
26:12 different elements of it than like the
26:14 giant tokamax but one that is very cool
26:16 is what what comes out of the reaction
26:19 is charged particles not heat almost all
26:21 other sort of like you know like a coal
26:23 plant or a natural gas plant or whatever
26:24 makes heat drives a steam turbine that’s
26:26 what it does helion makes charged
26:28 particles which push back on the magnet
26:30 and drive an electrical current down a
26:31 wire there’s no heat cycle at all and so
26:33 it can be a much simpler much more
26:35 efficient system and that I think is
26:37 like Mist out of the whole discussion on
26:39 Fusion but really great it also means we
26:42 don’t have to deal with like much
26:43 nuclear material we don’t ever have like
26:45 dangerous waste or even a dangerous
26:47 system you could like touch it pretty
26:48 shortly after it turns off has it proven
26:50 it’s thesis
26:52 we will have more to share there shortly
26:54 I think by like 2028 pending you know
26:58 good fortune with Regulators we could be
27:01 plugging them into the grid
27:03 I think we’ll do it you know a really
27:05 great demo well before that like
27:07 hopefully pretty soon is
28:03 foreign
, , , #Resumen #ENTREVISTA #Sam #Altman #CEO #OpenAI #GPT4 #ChatGPT #AGI , [agora]