Kind: captions
Language: en
I think compute is going to be the
currency of the future I think it will
be maybe the most precious commodity in
the world I expect that by the end of
this
decade
and
possibly somewhat sooner than that we
will have quite capable systems that we
look at and say wow that's really
remarkable the road to AGI should be a
giant power struggle I expect that to be
the case whoever builds AGI first gets a
lot of power
do you trust yourself with that much
power the following is a conversation
with Sam Altman his second time in the
podcast he is the CEO of open AI the
company behind GPT 4 Chad GPT Sora and
perhaps one day the very company that
will build
AGI this is Alex Freedman podcast to
support it please check out our sponsors
in the the description and now dear
friends here's Sam
Alman take me through the open AI board
Saga that started on Thursday November
16th maybe Friday November 17th for you
that was definitely the most painful
professional experience of my
life
and chaotic and
shameful and
upsetting and a bunch of other negative
things uh there were great things about
it too and I wish I wish it had not
been in such an adrenaline rush that I
wasn't able to stop and appreciate them
at the time but
um I came across this old tweet of mine
or this tweet of mine from that time
period which was like it was like you
know kind of going to your own eulogy
watching people say all these great
things about you and uh just like
unbelievable support from people I love
and care about
uh that was really nice um that whole
weekend I I kind of like felt with one
big exception I I felt like a great deal
of
love and very little
hate
um even though it felt like I just I
have no idea what's happening and what's
going to happen here and this feels
really bad and there were definitely
times I thought it was going to be like
one of the worst things to ever happen
for AI safety
well I also think I'm happy that it
happened relatively early I thought at
some point between when opening I
started and when we created AGI there
was going to be something crazy and
explosive that happened but there may be
more crazy and explosive things still to
happen um it
still I think helped us build up some
resilience and be ready
for more challenges in the future but
the thing you had a sense that you would
experience is some kind of power
struggle the road to AGI should be a
giant power struggle like the world
should I like well not should I expect
that to be the case and so you have to
go through that like you said iterate as
often as
possible uh in figuring out how to have
a board structure how to have
organization how to have um the kind of
people that you're working with how to
communicate all that in order to uh
deescalate the power struggle as much as
possible yeah pacify it but at this
point it
feels you know like something that was
in the past that was really unpleasant
and really difficult and painful but
we're back to work and things are so
busy and so intense that I don't spend a
lot of time thinking about it there was
a time after uh there was like this
Fugue State um for kind of like the
month after maybe 45 days after that
was I was just sort of like drifting
through the days I was so out of it um I
was feeling so down uh just on a
personal psychological level yeah really
painful um and hard to like have to keep
running open a ey in the middle of that
I just wanted to like crawl into to a
cave and kind of recover for a while but
you know now it's like we're just back
to working on the
mission well it's still useful to go
back there and
reflect on board structures on power
dynamics on how companies are run the
tension between research and product
development and money and all this kind
of stuff so that you whoever a very high
potential of building AGI would do so in
a slightly more organized less dramatic
way yeah in the future so there's value
there to go both the personal
psychological aspects of you as a leader
and also just the the board structure
and all this kind of messy stuff
definitely learned a lot about
um structure and incentives and um what
we need out of a a board um and I think
that is it is valuable that this
happened now in some sense um I think
this is probably not like the last high
stress moment of opening eye but it was
quite a high stress moment like company
very nearly got destroyed and we think a
lot
about many of the other things we've got
to get right for AGI but thinking about
uh how to build a resilient org and how
to build a structure that will stand up
to like a lot of pressure in the world
which I expect more and more as we get
closer I think that's super important do
you have a sense of how deep and
rigorous the deliberation process by the
board was like can you shine some light
on just human dynamics involved in
situations like this was it just a few
conversations and all of a sudden it
escalates and why don't we fire Sam kind
of thing I
think I think the board members were are
well meaning people on the whole um
and I believe
that in stressful situations
um where people feel time pressure or
whatever
uh people understandingly make
suboptimal decisions and I think one of
the challenges for open AI will be we're
going to have to have a board and a team
uh that are good at operating under
Under Pressure do you think the board
had too much power I think boards are
supposed to have a lot of power um but
one of the things that we did see is in
in most corporate structures boards are
usually answerable to shareholders you
know there's sometimes people have like
super voting shares or whatever um in
this case and I think one of the things
with our structure that we maybe should
have thought about more than we did is
that the board of a nonprofit has unless
you put other their rules in place like
quite a quite a lot of power they don't
really answer to anyone but themselves
and there's ways in which that's good
but what we'd really like is for the
board of open a to like answer to the
world as a whole as much as that's a
practical thing so there's a new board
announced yeah there's I
guess uh a new smaller board at first
and now there's a new Final board not a
final board yet we've added some We'll
add more added some okay
what is fixed in the new one that was
perhaps
broken in the previous one the old board
sort of got smaller uh over the course
of about a year it was nine and then it
went down to six and then we couldn't
agree on who to add and the board also
uh I think didn't have a lot of
experienced board members and a lot of
the new board members at open have just
have more experience as board members um
I think that'll help it's been
criticized some of the people that added
to the board I heard a lot of people
criticizing the addition of Larry
Summers for example what what's the
process of selecting the board like
what's involved in that so Brett and
Larry were kind of uh decided In the
Heat of the Moment over this like very
tense weekend and that was that mean
that weekend was like a real roller
coaster it's like a lot of a lot of ups
and downs um and we were trying to agree
on
new board members that both sort of the
executive team here and the old board
members felt would be reasonable um
Larry was actually one of their
suggestions the old board members um
Brett I think I had even previous to
that weekend suggested but he was you
know busy and didn't want to do it and
then we really needed help in Wood um we
talked about a lot of other people too
uh but that
was I felt like if I was going to come
back uh I needed new board members um I
didn't think I could work with the old
board again in the same configuration
although we then decided uh and I'm
grateful that Adam would stay um but we
wanted to get to uh we considered
various configurations decided we wanted
to get to a board of three and uh had to
find two new board members over the
course of sort of a short period of time
so those were decided honestly without
uh you know that's like you kind of do
that on the battlefield you don't have
time to design a rigorous process then
um for new board members since and new
board members will add going forward um
we have some criteria uh that we think
are important for the board to have
different expertise that we want the
board to have um unlike hiring an
executive where you need them to do one
role well the Board needs to do a whole
role of kind of governance and
thoughtfulness uh well and so one thing
that Brett says which I really like is
that you know we want to hire board
members in slates not as individuals one
at a time and uh you know thinking about
a group of people that will
bring nonprofit expertise expertise at
running companies sort of good legal and
governance expertise uh that's kind of
what we've tried to optimize for so is
technical Savvy important for the
individual board members not for every
board member but for certainly some you
need that that's part of what the Board
needs to do so I mean the interesting
thing that people don't understand about
open a I certainly don't is like all the
details of running the business when
they think about the board given the
drama they think about you they think
about
like if you reach AGI or you reach some
of these incredibly impactful products
and you build them and deploy them
what's the conversation with the board
like and they kind of think all right
what's the right Squad to have in that
kind of situation to deliberate look I
think you definitely need some technical
experts there and then you need some
people who are
like what can how can we deploy this in
a way that will help people in the world
the most and people who have a very
different perspective you know I think a
mistake that you or I might make is to
think that only the technical
understanding matters and that's
definitely part of the conversation you
want that board to have but there's a
lot more about how that's going to just
like impact society and people's lives
that you really want represented in
there too and you're just kind of are
you looking at the track record of
people or you're just having
conversations track record is a big deal
you of course have a lot of
conversations but I
um you know there's some roles where I
kind
of totally ignore track record and just
look at slope kind of ignore the Y
intercept thank you thank you for making
it mathematical for the for the audience
for a board member like I do care much
more about the Y intercept like I think
there is something deep to say about
track record there and experiences
sometimes very hard to replace do you
try to fit a polinomial function or
exponential one to the to the track
record that's not that it analogy
doesn't carry that far all right you
mentioned some of the low
points uh that weekend what were some of
the low points psychologically for you
uh did you consider going um to the
Amazon jungle and just taking IA and
disappearing forever or I mean there's
so many like it was that was a very bad
period of time there were great High
points too like uh my phone was just
like sort of non-stop blowing up with
nice messages from people I work with
every day people I hadn't talked to in a
decade I didn't get to like appreciate
that as much as I should have because I
was just like in the middle of this
firefight but that was really nice but
on the whole it was like a very painful
weekend and also just like a
very it was like a battle thought in
public to a surprising degree and that's
that was extremely exhausting to me much
more more than I expected um I think
fights are generally exhausting but this
one really was you know board did this
uh Friday afternoon I really couldn't
get much in the way of answers but I
also was just like well the board gets
to do this and
so I'm going to think for a little bit
about what I want to do but I'll try to
find the the blessing in disguise here
and I was like well
I you know my current job at opening eye
is or it was like to like run a you know
decently sized company at this point and
the thing I had always liked the most
was just getting to like work on work
with the researchers and I was like yeah
I can just go do like a very focused AI
research effort and I got excited about
that didn't even occur to me at the time
to like possibly that this was all going
to get undone this was like Friday
afternoon so you've accepted your the
death of very quickly like within you
know I mean I went through like a little
period of confusion and rage but very
quickly and by Friday night I was like
talking to people about what was going
to be next and I was excited about that
um I think it was Friday night evening
for the first time that I heard from the
exec team here which is like Hey we're
going to like fight this and you know we
think whatever and then I went to bed
just still being like okay excited like
onward were you able to sleep not a lot
it was one of one of the weird things
was it was this like period of four four
and a half days days where sort of
didn't sleep much didn't eat much and
still kind of had like a surprising
amount of energy it was you learned like
a weird thing about adrenaline in more
time so you kind of accepted the death
of a you know this baby opening I was
excited for the new thing I was just
like okay this was crazy but whatever
it's a very good coping mechanism and
then Saturday morning uh two of the
board members called and said hey we you
know destabilize we didn't mean to
destabilize things we don't want to
destroy a lot of value here you know can
we talk about you coming back and I
immediately didn't want to do that but I
thought a little more and I was like
well I you don't really care about the
people here the partners shareholders
like all of the I love this company and
so I thought about it and I was like
well okay but like here's here's the
stuff I would need and and then the most
painful time of all was over the course
of that weekend
um I kept thinking and being told and we
all kept not just me like the whole team
here kept thinking while we were trying
to like keep and I
stabilized while the whole world was
trying to break it apart people trying
to recruit whatever um we kept being
told like all right we're almost done
we're almost done we just need like a
little bit more time um and it was this
like very confusing State and then
Sunday evening when again like every few
hours I expected that we were going to
be done and we're going to like figure
out a way for me to return and things to
go back to how they were um the board
then uh appointed a new interim
CEO and then I was like I mean that is
that is that feels really bad that was
the low point of the whole
thing you know I'll tell you something I
it felt very painful but I felt a lot of
love that whole weekend it was not other
than that one moment Sunday night I
would not characterize my emotions as
anger or hate um but I really just
like I felt a lot of love from people
towards
people it was like painful but it would
like the dominant emotion of the weekend
was love not hate you've spoken highly
of uh Mera moradi that she helped
especially as he put in a tweet in The
Quiet Moments When it counts perhaps we
could take a bit of a tangent what do
you admire about well she did a great
job during that weekend in a lot of
chaos but but people often see leaders
in the moment in like the crisis moments
good or
bad um but a thing I really value and
leaders is how people act on a boring
Tuesday at 9:46 in the morning and in in
just sort of the the the normal drudgery
of the
day-to-day how someone shows up in a
meeting the quality of the decisions
they make that was what I meant about
the Quiet Moments meaning like most of
the work is done on a day by day in the
meeting by meeting just just be present
and and make great decisions yeah I me
mean look what you wanted to have wanted
to spend the last 20 minutes about and I
understand is like this one very
dramatic weekend yeah but that's not
really what opening eye is about opening
eye is really about the other seven
years well yeah human civilization is
not about the invasion of the Soviet
Union by Nazi Germany but still that's
something people focus on because very
understandable it gives us an insight
into human nature the extremes of human
nature and perhaps some of the damage
and some of the triumphs of human
civilization can happen in those moments
it's like
illustrative let me ask you about
Ilia is he being held hostage in a
secret nuclear facility no what about a
regular secret facility no what about a
nuclear non secret facility neither not
that either I mean this becoming a meme
at some point you've known Ilia for for
a long time he was obviously in part
part of this drama with the board and
all that kind of
stuff what's your relationship with him
now I love Ilia I have tremendous
respect for Ilia I uh I don't have
anything I can like say about his plans
right now that's that's a question for
him um but I really hope we work
together for you know certainly the rest
of my
career he's a little bit younger than me
maybe he works a little bit
longer you know there's a there's a meme
that he saw something like he maybe saw
AGI and that gave him a lot of worry
internally uh what did ilas
see uh oh has not seen AGI none of us
have seen AGI we've not built
AGI
uh I do think uh one of the many things
that I really love about Ilia is he
takes AGI and the safety concerns
broadly speaking you know including
things like the impact this is going to
have on society very seriously and we as
we continue to make significant progress
um Ilia is one of the people that I've
spent the most time over the last couple
of years talking about what this is
going to mean what we need to do to
ensure we get it right to ensure that we
succeed at the mission um
so Ilia did not see AGI um but Ilia is
a credit to humanity in terms of how
much he
thinks and worries about making sure we
get this right I've had a bunch of
conversation with him in the past I
think when he talks about technology
he's always like doing this long-term
thinking type of thing so he's not
thinking about what this is going to be
in a year he's thinking about in 10
years yeah just thinking from first
principles like
okay if the scales what are the
fundamentals here where is this going
and so that that's a foundation for them
thinking about like all the other safety
concerns and all that kind of stuff um
which makes him really fascinating human
uh to talk with do you have any idea why
he's been kind of quiet is it he's just
doing some soul searching again I don't
want to like speak for oh yeah I think
that you should ask him that
um he's definitely a thoughtful guy
uh I think I kind of think ailia is like
always on a soul search in a really good
way yes yeah also he appreciates the
power of Silence also I'm told he can be
a silly guy which I've never I've never
seen that side of it's very sweet when
that
happens I've never witnessed a silly
Ilia but um I look forward to to that as
well I was at a dinner party with him
recently and he was playing with a puppy
and I and he was like in a very silly
moood very endearing and I was thinking
like oh man this is like not the side of
the ILO that the world sees the most so
just to wrap up this whole Saga are you
feeling good about the board structure
about all of this and like where it's
moving I feel great about the new board
in terms of the structure of openi I you
know one of the board's tasks is to look
at that and see where we can make it
more robust um we wanted to get new
board members in place first uh but you
know we clearly learned a lesson about
structure throughout this process I
don't have I think super deep things to
say it was a crazy very painful
experience I think it was like a perfect
storm of weirdness it was like a preview
for me of what's going to happen as the
stakes get higher and higher in the need
that we have like robust governance
structures and process as in people
um I am kind of happy it happened when
it did but it
was a shockingly painful thing to go
through did it make you be more hesitant
and trusting people yes just in a
personal level I think I'm like an
extremely trusting person I have always
had a life philosophy of you know like
don't worry about all of the paranoia
don't worry about the edge cases you
know you get a little bit screwed in
exchange for getting to live with your
guard down and this was so shocking to
me I was so caught off guard that it has
definitely
changed and I really don't like this
it's definitely changed how I think
about just like default Trust of people
and planning for the bad scenarios you
got to be careful with that are you
worried about becoming a little too
cynical um I'm not worried about
becoming too cynical I think I'm like
the extreme opposite of a cynical person
but I'm I'm I'm worried about just
becoming like less of a default trusting
person I'm actually not sure which mode
is best to operate in for a person who's
developing
AGI trusting or untrusting it's an
interesting Journey you're
on but in terms of structure see I'm
more interested on the human level like
how do you surround yourself with humans
that're building cool but also are
making wise decisions because the more
money you start making the more power
the thing has the weirder people get you
know I think you could like you can make
all kinds of comments about
the board members and the level of trust
I should have had there or how I should
have done things differently but in
terms of the team here I think you'd
have to like give me a very good grade
on that one um and I have uh just like
enormous gratitude and trust and respect
for the people that I work with every
day and I think being surrounded with
people like that
is is really
[Music]
important our mutual friend Elon sued
open
AI m is the essence of what he's
criticizing to what degree does he have
a point to what degree is he wrong I
don't know what it's really about we
started
off just thinking we're going to be a
research lab and having no idea about
how this technology was going to go it's
hard to because it was only you know
seven or eight years ago it's hard to go
back and really remember what it was
like then but this before language
models were a big deal this was before
we had any idea about an API or selling
access to a chatbot is before we had any
idea we were going to productize it all
so we're like we're just like going to
try to do research and you know we don't
really know what we're going to do with
that I think with like many new
fundamentally new things you start
fumbling through the dark and you make
some assumptions most of which turn out
to be wrong and then it became clear
that we were going to need to
do different things and also have huge
amounts more Capital so we said okay
well the structure doesn't quite work
for that how do we patch the structure
um and then you patch it again and Patch
it again and you end up with something
that does look kind of eyebrow raising
to say the least but we got here
gradually with I think reasonable
decisions at each point along the way
and doesn't mean I wouldn't do it
totally differently if we could go back
now with an oracle but you don't get the
Oracle at the time but anyway in terms
of what elon's real motivations here are
I don't
know to the degree you remember what was
the response that open AI gave in the
blog post can you summarize
it oh we just said
like you know Elon said this set of
things here's our character ation or
here's the sort of not our
characterization here's like the
characterization of how this went down
um we tried to like not make it
emotional and just sort of say
like here's the history I do
think there's a degree
of mischaracterization from Elon here
about one of the points he just made
which is the degree of uncertainty has
at the time you guys are a bunch of like
a small group of
researchers craz talking about AGI when
everybody's laughing at that thought
wasn't that long ago Elon was crazily
talking about launching Rockets yeah
when people were laughing at that
thought
uh so I think he'd have more empathy for
this I mean I I do think that there's
personal stuff here that there was a
split that open Ai and a lot of amazing
people here chose the part ways of Elon
so there's a personal Elon chose the
part
ways can you describe that exactly the
the the choosing to part ways he thought
open ey was going to fail um he wanted
total control to sort of turn it around
we wanted to keep going in the direction
that now has become open AI he also
wanted Tesla to be able to build an AGI
effort at various times he wanted to
make open AI into a for-profit company
that he could have control of or have it
merge with Tesla um we didn't want to do
that and he decided to leave which
that's
fine so you're saying and that's one of
the things that the blog post says is
that he wanted open AI to be basically
acquired by Tesla yeah in the same way
that or maybe something similar or maybe
something more dramatic than the
partnership with Microsoft my memory as
the proposal was just like yeah like get
acquired by Tesla and have Tesla have
full control over it I'm pretty sure
that's what it was so what is the word
open in open AI mean to Elon at the time
Ilia has talked about this in the email
exchanges and all this kind of stuff
what does it mean to you at the time
what does it mean to you now I would
definitely pick a diff speaking of going
back with an oracle I'd pick a different
name
um one of the things that I think
opening eye is doing that is the most
important of everything that we're doing
is putting powerful technology in the
hands of people for free as a public
good not we're not you know we don't run
ads on our free version we don't
monetize it in other ways we just say
it's part of our mission we want to put
increasingly powerful tools in the hands
of people for free and get them to use
them and I
think that kind of open is really
important to our mission I think if you
give people great tools and teach them
to use them or don't even teach them
they'll figure it out and let them go
build an incredible future for each
other with that uh that's a big deal so
if we can keep putting like free or
lowcost or free and low cost powerful AI
tools out in the world uh I it's a huge
deal for how we fulfill the mission um
open source or not yeah I think we
should open source some stuff and not
other stuff uh the it does become this
like religious battle line where Nuance
is hard to have but I think Nuance is
the right answer so he said change your
name to closed Ai and I'll drop the
lawsuit I mean is it going to become
this
Battleground in in the land of memes
above I think that
speaks to the seriousness with which
Elon means the
lawsuit and
uh yeah I mean that's like an
astonishing thing to say I think like
well I don't think the lawsuit may maybe
correct me if I'm wrong but I don't
think the lawsuit is legally serious
it's more to make a point about the
future of AGI and the company that's
currently leading the
way so look I mean grock had not open
sourced anything until people pointed
out it was a little bit hypocritical and
then he announced that Gro will open
source things this week I don't think
open source versus not is what this is
really about for him well we'll talk
about open source and not I do think
maybe criticizing the competition is
great just talking a little that's
great but friendly competition versus
like
I personally hate lawsuits yeah look I
think this whole thing is like
Unbecoming of a builder and I respect
Elon as one of the great Builders of our
time and
um I know he knows what it's like to
have like haters attack him and it makes
me extra sad he's doing it toss yeah
he's one of the greatest Builders of all
time potentially the greatest builder of
all time it makes me sad and I think it
makes a lot of people sad like there's a
lot of people who've really looked up to
him for a long time and
said this I said you know in some
interview or something that I missed the
old Elon and the number of messages I
got being like that exactly encapsulates
how I feel I think he should just win he
should just make X grock beat GPT and
then GPT beats Croc and it's just a
competition and it's it's beautiful for
everybody but on the question of Open
Source do you think there's a lot of
companies playing with this idea it's
quite interesting I would
say
surprisingly has led the way on this or
like uh at least took the first step in
the game of chess of like really open
sourcing the model of course it's not
the state-ofthe-art model but open
sourcing
llama and you Google is flirting with
the idea of open sourcing a smaller
version have you what are the pros and
cons of open sourcing have you played
around this idea yeah I think there
there is definitely a place for open
source models particularly smaller
models that people can run locally I
think there's huge demand for um I think
there will be some open source models
there will be some close Source models
uh this it won't be unlike other
ecosystems in that way I listened to uh
all-in podcast talking about this this
lwuit and all that kind of stuff and
they were more concerned about the
precedent of going from nonprofit to
this cap for
profit what president ass says for other
startups is that I don't I would heavily
discourage any startup that was thinking
about starting as a nonprofit and adding
like a for-profit arm later I'd heavily
discourage them from doing that I don't
think we'll set a precedent here okay so
most most startups should go just for
sure and again if we knew what was going
to happen we would have done that too
well like in theory if you like dance
beautifully here you could there's like
some tax incentives or whatever but I I
don't think that's like how most people
think about these things just not
possible to save a lot of money for a
startup if you do it this way no I think
there's like laws that would make that
pretty difficult where do you hope this
goes with
Elon what this this tension this dance
what do you hope this like if we go one
two 3 years from
now your relationship with him on a
personal level too like friendship
friendly competition just all this kind
of
stuff yeah I mean I really respect Elon
um
and I hope that years in the future we
have an amicable
relationship yeah I hope you guys have
an amicable relationship like this
month and just compete and win and uh
and explore these ideas together um I do
suppose there's competition for talent
or whatever
but it should be friendly competition
just build build cool
and Elon is pretty good at building cool
but so are you so speaking of cool
uh Sora there's like a million
questions I could ask first of all it's
amazing it truly is amazing on a product
level but also just on a philosophical
level so let me just a technical
philosophical ask what do you think it
understands about the
world more or less than GPT 4 for
example like the world model when you
train on these patches versus language
tokens I think all of these models
understand something more about the
world model than most of us give them
credit for and because they're also very
clear things they just don't understand
or don't get right it's easy to like
look at the weaknesses see through the
veil and say ah this is just this is all
fake but it's not all fake it's just
some of it works and some of it doesn't
work like I remember when I started
first watching Sora videos and I would
see like a person walk in front of
something for a few seconds and include
it and then walk away and the same thing
was still there I was like oh it's
pretty good or there's examples where
like the underlying physics looks so
well represented over you know a lot of
steps in a sequence it's like oh this is
this is like quite impressive but like
fundamentally these models are just
getting better and that will keep
happening if you look at the trajectory
from Dolly 1 to 2 to 3
Sora you know there were a lot of people
that were dunked on each version saying
it can't do this it can't do that and
like look at it now so well the thing
you just mentioned is kind of with the
occlusions is basically modeling the
physics of the threedimensional physics
of the world sufficiently well to
capture those kinds of things well or
like under or yeah maybe you can tell me
in order to deal with occlusions what
does the world model need to yeah so
what I would say is it's doing something
to deal with occlusions really well what
I represent that it has like a great
underlying 3D model of the world it's a
little bit more of a stretch but can you
get there through just these kinds of
two-dimensional training data
approaches uh it looks like this
approach is going to go surprisingly far
I don't want to speculate too much about
what limits it will surmount and which
it won't but what are some interesting
limitations of the system that you've
seen I mean there's been some fun ones
you've posted there's all kinds of one I
mean like you know cats sprout in a
extra limb at random points in a video
uh like pick what you want but there's
still a lot of problem a lot of
weaknesses do you think it's a
fundamental flaw of the
approach or is it just you know bigger
model or better like technical details
or better data more data is going to
solve those the cat sprouting I say yes
to both like I think there is something
about the approach which just seems to
feel different from how we think and
learn and whatever
and then also I think it'll get better
with skill like I mentioned llms have
tokens text tokens and Sora has visual
patches so it converts all visual data
at diverse kinds of visual data videos
and images into patches is the training
to the degree you can say fully
self-supervised there's there some
manual labeling going on like what's the
involvement of humans in all
this I mean without saying anything
specific about the Sora approach we we
use lots of human data
in our
work but not internet scale data so lots
of humans Lots is a complicated word Sam
I think Lots is a fair word in this
case but it doesn't because to me lots
like listen I'm an introvert and when I
hang out with like three people that's a
lot of people four people that's a lot
but I suppose you mean more than more
than three people work on labeling the
data for these models yeah okay all
right but fundamentally there's a a lot
of self-supervised learning cuz what you
mentioned in the technical report is
internet scale data that's another
beautiful it's like poetry uh so it's a
lot of data that's not human label it's
like it's selfs supervised in that way
yeah and then the question is how much
how much data is there on the internet
that could be used in this that uh is
conducive to this kind of
self-supervised way if only we knew the
details of the self-supervised do you
have you considered opening it up a
little more details we have you mean for
Sora specifically Sora specifically the
because it's so
interesting that like can this L can the
same magic of llms now start moving
towards visual data and what does that
take to do that I mean it looks to me
like yes but we have more work to do
sure what are the dangers why are you
concerned about releasing the system
what uh what are some possible dangers
of this I Frankly Speaking one thing we
have to do before releasing the system
is is just like get it to work
at a level of efficiency that will
deliver the scale people are going to
want from this so that I don't want to
like downplay that and there's still a
ton ton of work to do there but you know
you can imagine
like issues with deep fakes
misinformation um like we try to be
thoughtful company about what we put out
into the world and it doesn't take much
thought to think about the ways this can
go badly there's a lot of tough
questions here uh you're dealing in a
very tough space do you think training
AI should be or is fair use under
copyright law I think the question
behind that question is do people who
create valuable data deserve to have
some way that they get compensated for
use of it and that I think the answer is
yes I don't know yet what the answer is
people have proposed a lot of different
things we've some tried some different
models but you know if I'm like an
artist for
example a I would like to be able to opt
out of people generating art in my style
and B if they do generate art in my
style I'd like to have some economic
model associated with that yeah it's
that uh transition from CDs to Napster
to
Spotify we have to figure out some kind
of model the model changes but people
have got to get paid well there should
be some kind of intive if we zoom out
even more for humans to keep doing cool
everything I worry about humans are
going to do cool and Society is
going to find some way to reward it I I
that seems pretty hardwired we want to
create we want to be useful we want to
like achieve status in whatever way
that's not going anywhere I don't think
but the reward might not be monetary
Financial it might be like Fame and
celebration of other cool maybe
Financial in some other way I guess I
don't think we've seen like the last
evolution of how the economic system is
going to work yeah but artists and
creators are worried when they see Sora
they're like holy sure artists were
also super worried when photography came
out yeah and then photography became a
new art form and people made a lot of
money taking pictures
and I think things like that will keep
happening people will use the new Tools
in new ways if we just look on YouTube
or something like this how much of that
will be using Sora like
AI generated content do you think in the
next five years people talk about like
how many jobs is they going to do in
five years and and the framework that
people have is what percentage of
current jobs are just going to be
totally replaced by some AI doing the
job the way I think about it is not what
percent of jobs AI will do but what
percent of tasks will AI do and over
what time Horizon so if you think of all
of the like five second tasks in the
economy five minute tasks the five hour
tasks maybe even the five day tasks how
many of those can AI do and I think
that's a way more interesting impactful
important question
than how many jobs AI can do because it
is a tool that will work at increasing
levels of sophistication and over longer
and longer time
Horizons for more and more tasks and let
people operate at a higher level of
abstraction so maybe people are way more
efficient at the job they do and at some
point that's not just a quantitative
change but that's a qualitative one too
about
the kinds of problems you can keep in
your head I think that for videos on
Youtube it'll be the same many videos
maybe most of them will use AI tools in
the production but they'll still be
fundamentally driven by a person
thinking about it putting it together
you know doing parts of it sort of
directing and running it yeah it's so
interesting I mean it's scary but it's
interesting to think about I tend to
believe that humans like to watch other
humans or other human hum really care
about other humans a lot yeah if there's
a cooler thing that's more that's better
than a
human humans care about that for like
two days and then they go back to humans
that seems very deeply wired it's the
whole chess thing oh yeah but no let's
everybody keep playing CH and Let's
ignore the elephant in the room that
humans are really bad at chess relative
to AI systems we still run races and
cars are much faster I mean this is
there's like a lot of examples yeah and
maybe just be
tooling like in the Adobe sweet type of
way where you can just make videos much
easier and all that kind of
stuff listen I hate being in front of
the camera if I can figure out a way to
not be in front of the camera I would
love it unfortun it'll take a while like
that generating faces it's it's getting
there but generating faces in video
format is tricky when it's specific
people versus generic people let me ask
you about gbt
4 so many questions uh first of all also
amazing it's looking back it'll probably
be this kind of historic pivotal moment
with 3 five and four which had BT maybe
five will be the pivotal moment I don't
know hard to say that looking forwards
we never know that's the annoying thing
about the future it's hard to predict
but for me looking back GPT for Chad gbt
is pretty damn impressive like
historically impressive so allow me uh
to ask ask what's been the most
impressive capabilities of gp4 to you
and gp4
turbo I think it kind of sucks H typical
human also gotten used to an awesome
thing no I think it is an amazing thing
um
but relative to where we need to get to
and where I believe we will get to uh
you know at the time of like
gpt3 people were like oh this is amazing
this is this like Marvel of technology
and it is it was uh but you know now we
have gp4
and look at GB3 and you're like that's
unimaginably horrible um I expect that
the Delta between five and four will be
the same as between four and three and I
think it is our job to live a few years
in the future and remember that the
tools we have now
are going to kind of suck looking
backwards at them
and that's how we make sure the future
is better what are the most glorious
ways in that GPT for sucks meaning uh
what are the best things it can do what
are the best things it can do and the
the limits of those best things that
allow you to say it sucks therefore
gives you an inspiration and hope for
the future you know one thing I've been
using it for more recently is sort of a
like a brainstorming partner Y and for
that there's a glimmer of something
amazing in there I don't think it gets
you know when people talk about it it
what it does they're like ah it helps me
code more productively it helps me write
more faster and better it helps me you
know translate from this language to
another all these like amazing things
but there's something about the like
kind of creative brainstorming partner I
need to come up with a name for this
thing I need to like think about this
problem in a different way I'm not sure
what to do here
uh that I think like gives a glimpse of
something I hope to see more of um one
of the other things that you can see
like a very small glimpse of
is when it can help on longer Horizon
tasks you know break down some multiple
steps maybe like execute some of those
steps search the internet write code
whatever put that together uh when that
works which is not very often it's like
very
magical the iterative back and forth
with a human well it works a lot for me
what do you mean it uh iterative back
and forth the human can get more often
when it can go do like a 10-step problem
on its own oh doesn't work for that too
often sometimes at multiple layers of
abstraction or do you mean just
sequential both like you know to break
it down and then do things at different
layers of abstraction and put them
together look I don't want to I don't
want to like downplay the accomplishment
of gp4
um but I don't want to overstate it
either and I think this point that we
are on an exponential curve we will look
back relatively soon at gp4 like we look
back at gpt3 now that said I mean Chad
gbt was a transition to where people
like started to believe it there was a
kind of there is an uptick of believing
not internally at open AI perhaps
there's Believers here but when you
think and in that sense I do think it'll
be a moment where a lot of the world
went from not believing to believing um
that was more about the chat gbt
interface than the and and by the
interface and product I also mean the
post training of the model and how we
tune it to be helpful to you and how to
use it than the underlying model itself
how much
of those two uh each of those things are
important the underlying model and the
rlf or something of that nature that
Tunes it to be more compelling to the
human more uh effective and productive
for the human I mean they're they're
both super important but the the the rhf
the post-training step the you know
little wrapper of things that from a
compute perspective little wrapper of
things that we do on top of the base
model even though it's a huge amount of
work that's really important to say
nothing of the product that we build
around it
um you know in some sense like we did
have to do two things we had to invent
the underlying technology and then we
had to figure
out
how to make it into a product people
would
love which is not just about the actual
product work itself but this whole other
step of how you align it and make it
useful and how you make the scale work
where a lot of people can use it at the
same time all that kind of stuff and
that but you know that was like unnown
difficult thing like we knew we were
going to have to scale it up we had to
go do two things that had like never
been done before uh that were both like
I would say quite significant
achievements and then lot of things like
scaling it up that other companies have
had to do
before how does the the context window
of going from 8K to 128k tokens compare
from the from GPT 4 to to GPT 4 Turbo
people like long most people don't need
all the way to 128 most of the time
although you know if we dream into the
distant future we'll have like like way
distant future we'll have like context
length of several billion you will feed
in all of your information all of your
hisory over time and it'll just get to
know you better and better and that'll
be great for now uh the way people use
these models they're not doing that and
you know people sometimes Post in a
paper or you know a significant fraction
of a code repository whatever
um but most usage of the models is not
using the long context most of the time
I like that this is year I Have a Dream
speech one day you'll be judged by the
full context of your character or of
your whole lifetime that's interesting
so like that's part of the expansion
that you're hoping for is a greater and
greater context there was this I saw
this internet clip once I'm going to get
the numbers wrong but it was like Bill
Gates talking about the amount of memory
on some early
computer maybe 64k maybe 640k something
like that and most of it was used for
the screen
buffer and he just couldn't seemed
genuine this couldn't imagine that the
world would eventually need gigabytes of
memory a computer or terabytes memory in
a
computer
um and you always do or you always do
just need to like follow the exponential
of technology and and we're going to
like we will find out how to use better
technology so I can't really imagine
what it's like right now for context
links to go out to the billion someday
and they might not literally go there
but effectively it'll feel like that
um but I know we'll use it and really
not want to go back once we have it yeah
even saying billions 10 years from now
might seem dumb because it'll be
like trillions upon trillions sure
there'd be some kind of
breakthrough that will effectively feel
like infinite context but even 120 I
have to be honest I haven't pushed it to
that degree maybe putting in entire
books or like parts of books and so on
papers what are some interesting use
cases of GPT 4 that you've seen the
thing that I find most interesting is
not any particularly case that we can
talk about those but it's people who
kind of
like this is mostly younger people but
people who use it as like their default
start for any kind of knowledge work
task yeah and it's the fact that it can
do a lot of things reasonably well you
can use gptv you can use it to help you
write code you can use it to help you do
search you can uh use it to like edit a
paper the most interesting thing to me
is the people who just use it as the
start of their workflow I do as well for
for many things like uh I use it as a u
a partner for reading
books it helps me think help me think
through ideas especially when the books
are classic so it's really well written
about and it actually is
is I I find it often to be significantly
better than even like Wikipedia on
well-covered topics it's somehow more
balanced and more nuanced or maybe it's
me but it inspires me to think deeper
than a Wikipedia article does I'm not
exactly sure what that is you mentioned
like this collaboration I'm not sure
where the magic is if it's in here or if
it's in there or if it's somewhere in
between not sure uh but one of the
things that concerns me for knowledge
task when I start with GPT is I'll
usually have to do fact checking
after like check that it didn't come up
with fake stuff how how do you figure
that out that you know GPT
can come up with fake stuff that sounds
really convincing so how do you ground
it in truth that's obviously an area of
intense interest for us uh I think it's
going to get a lot better with upcoming
versions but we'll have to work on it
and we're not going to have it like all
solved this year well the scary thing is
like as it gets better you'll start not
doing the factchecking more and more
right I I'm of two minds about that I
think people are like much more
sophisticated users of Technology than
we often give them credit for and people
seem to really understand that GPT any
of these models hallucinate some of the
time and if it's mission critical you
got to check it except journalists don't
seem to understand that I've seen
journalists half acidly just using GPT
for it's of the long list of things I'd
like to dunk on journalists for this is
not my top criticism of them well I
think the bigger criticism is perhaps
the pressures and the incentives of
being a journalist is that you have to
work really quickly and this is a
shortcut I I would love our society to
incentivize like I would too long like a
journal
journalistic efforts that take days and
weeks and and rewards great in-depth
journalism also journalism that presents
stuff in a