February 20, 2025 • 10 mins
In this week's monologue, Ed Zitron walks you through the early days of OpenAI - and the simmering war between Elon Musk and Sam Altman over its non-profit status.
Suit: https://www.courtlistener.com/docket/69013420/musk-v-altman/
---
LINKS: https://www.tinyurl.com/betterofflinelinks
Newsletter: https://www.wheresyoured.at/
Reddit: https://www.reddit.com/r/BetterOffline/
Discord: chat.wheresyoured.at
Ed's Socials:
https://www.instagram.com/edzitron
https://bsky.app/profile/edzitron.com
https://www.threads.net/@edzitron
See omnystudio.com/listener for privacy information.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Speaker 1 (00:01):
Aw Zone Media. Hello On, Welcome to this week's Better
Offline Monologue. I'm your host ed ZiT Trump, and I
know some of you are going to say, Ed, didn't
(00:22):
you say we'd get a second part? Didn't you say
we get a second part? Ed? Where's the second part?
It's coming tomorrow. You get a monologue as well. Good Lord.
The complaints from some of you just kidding. You're all
very nice. Now, because I deeply hate myself, I decided
to sit down and read case four to two four
CV zero four seven two to two ygr from the
United States District Court of West the Northern District to California. Nevertheless,
(00:46):
what I'm talking about, of course, is Elon Musk's lawsuit
against Open a I, filed in August. November of last year,
there was an amended complaint. Nevertheless, Elon Musk is alleging
multiple kinds of fraud, as well as violations of the
Sherman Act and to monopoly law from the late eighteen hundreds,
which most notably was the same law that might lead
to the breakup of Google's ad taken search businesses. Elon Musk,
(01:08):
he's suing, and he loves to sue, but in Layman terms.
Musk alleges that Sam Wortman tricked him into funding open
ai as a charity when he actually wanted it to
be more like a for profit entity, a classical start
up model. Musk also alleges a conspiracy by open ai
to stop people who invested in open ai from investing
in other generative AI companies, specifically Musk zons Ai as
(01:30):
long as others like Anthropic. This sounds like some Musky
and bullshit, but this is actually true. It was reported
by The Information and other outlets. The lawsuit itself is contrived,
including annoying things like Musk's lawyers referring to open AI's
tax exempt nonprofit as a for profit market paralyzing Gorgan
Just you don't need to write like this. He fucking
(01:51):
loses anyway. It is pretty interesting, though, and it explores
the deeply weird beginnings of open ai itself. To explain
Joe and was originally founded in twenty fifteen by Elon Musk,
Sam Mortman and a selection of other engineers specifically as
a non profit making open source artificial intelligence, and it
was meant to be a research house. Now another thing
(02:13):
is it was specifically made as a reaction to Google's
acquisition of artificial intelligence firm deep Mind. The plan, according
to emails shared as part of the lawsuit, was to
beat Google to the punch by making artificial general intelligence,
you know, the entirely fictional concept of a conscious autonomous computer,
and then they go and open source it in what
Sam Moortman called an AI Manhattan project. Just I could
(02:36):
go into the history there, but is that really what
you want to compare this anyway? Anyway? Aortmand would go
on to tell Elon Musk that the mission would be
to create the first general AI AGI and use it
for an individual empowerment i e. The distributed version of
the future that seems the safest. More generally, safety should
be a first class requirement. And that is a quote,
(02:58):
by the way, with the technol owned by the foundation
referring to open AI and use for the gift of
the world. Just a lot of bollocks really, Anyway. Things
began to get tense in September twenty sixteen when Sam
Wortman arranged a deal with Microsoft to buy sixty million
dollars of Compute for Well for ten million dollars in
exchange for evangelizing Microsoft as zero as their preferred cloud provider,
(03:21):
along with some sort of vague consultancy services over Microsoft's models.
Musk would respond to the terms by saying, fine with me.
If they don't use the active messaging would be worth
way more than fifty million. Nots seemed like a Microsoft's
marketing bitch. Two months later, Microsoft were put out a
blog post saying that open ai was choosing as zero
as their primary cloud platform and that open ai would
(03:43):
become an early adopter of as zure En series virtual machines,
some of Microsoft's early GPU compute instances. It's been going
quite a while. A year later, in an exhibit from
the trial from September twenty to twenty seventeen, things would
get a little more fraid with Iliasuitskava, a gifted engineer
recruited by Musk in AI's earliest days, sending an email
(04:03):
to both Musk and Aortman sharing concerns about the future.
Aortman worried about how much money would cost the fund.
Open Ai had been considering finding a way to make
it a wouldn't you guess it? For profit entity, but
Sitzkev had other problems and was far more worried about
Altman and Musk. In the email, Suitzkeave of raised concerns
that Elon Musk wanted unilateral absolute control over the AGI,
(04:27):
and that while Musk had claimed otherwise in negotiating how
to keep open Ai going, it was very clear that
and I quote absolute control was extremely important to him.
As an example, Suitzkeaver added that Musk had said that
he needed to be CEO of the new company so
that everyone would know that he was the one in charge,
even though he also stated that he hated being CEO
and would much rather not be CEO. Suitzkeva added the
(04:51):
Musk's concerns that there would be an AGI dictatorship run
by Demis Hassabis, CEO of Deep Mind, but that in
the current structure that Musk was suggest he would become
a dictator if he chose to all very good stuff.
Siitzkav bizarrely then immediately moved on to say something very
very similar to Sam Altman, saying that and I quote
he didn't understand why the CEO title was so important,
(05:14):
and that Sam Altman's reasons had changed, and that it
was really hard to understand what was driving them Siitzkev
also added a question and I quote, is AGI truly
your primary motivation? How does it connect to your political goals?
How has your thought process changed over time? Ortmand would
reassure Musk, both personally and through others, that he remained
(05:35):
focused on open AI's nonprofit mission. In January twenty eighteen,
Orton would suggest a ridiculous idea selling cryptocurrency to fund
open Ai, which Musk would warn would simply result in
a massive loss of a credibility for open Ai everyone
associated with the ico, referring of course, to an initial
coin offering, a flimsy idea that just means just buy
(05:55):
a bunch of tokens before the thing goes live, basically
how crypto works. I guess it was a whole boom.
I'm not doing a fucking podcast about it. Let's move on.
Musk would step down from open ai in February twenty eighteen,
and a month later, Sam Wonan would propose a fixed
maximum term equity race, essentially selling stock in open ai
but an associated entity. Yet it was still a nonprofit
(06:20):
at the time, and that had a maximum amount you
could make on buying it. It's just very confusing and
What it basically means is it means that they would
create an entity on the side that you could raise
money for that would also own all the bits. I'll
get to that in a second. Nevertheless, this is all
extremely dodgy and weird. Around a year later, in twenty nineteen,
(06:43):
Sam Wonman would eventually create the legally precarious for profit
arm of open Ai what I was just talking about,
and it was called open AILP and immediately, according to
Elon Musk's lawsuit, transferred most of the company's assets and stuff.
The same year, open Ai would strike an exclusive partnership
with Microsoft to provide the compute for their models. As
part of the deal, open Ai would give Microsoft full
(07:05):
license to use their pre AGI intellectual property and research,
which is to say, literally everything they've ever made, and
this would in turn make well, this is the funny,
weird part. This is the really crazy. This is the
part that really gets me. They would own everything. Microsoft
would own everything until they hit AGI. Now, AGI at
(07:26):
this point has been defined by open Ai and Microsoft
as when they hit one hundred billion dollars in profit.
Every time I read about and talk about this stuff,
I just think, who is the idiot? Here, is it
Satching the Dellar, Is it Sam Altman or they both
just the kind of mediocre rich guy who just bounces
their skulls together and they say who has the shittiest idea?
(07:47):
Who will be the dumbest boy today. Nevertheless, Microsoft owns
everything open Ai makes until they invent Agi, by which
I mean they make one hundred billion dollars in profit.
It's all so goddamn stupid. It's also stupid. Now at
some point I want to do an entire episode on
this lawsuit because it's got so many exhibits and so
(08:08):
many warring incentives. Elon Musk's Xai competes directly with open
ai to make large language models that no one really
needs and that cost more to run than they will
ever make. And this lawsuit, as with others, features broad
demands for discovering depositions of people at LinkedIn co founder
and former open ai board member Read Hoffman, and attempts
(08:29):
to name both Microsoft and Hoffman himself as co defendants.
Since filing the lawsuit, and the Elon Musk led consortium
of Bias has offered ninety seven point four billion dollars
for the assets of open AI's charity, an often that
would require multiple different government agencies to approve, which open
Aies board has now declined either way. While Musk is
regularly full of ship, he's right about one thing. Sam
(08:52):
Woman clearly had no intention of ever keeping open ai
as nonprofit, nor was he ever dedicated to doing so,
or really anything of them make in himself CEO and
getting a billion dollars. Since twenty nineteen, open ai has
raised over twenty billion dollars in funding, and it's reportedly
raising as much as forty billion dollars in the next round,
led by fucking Masayoshi son of soft Bank. It's so good.
(09:14):
I love it. And they're likely doing this because the
company burned five billion dollars in twenty twenty four, and
it is said to as much as double that in
twenty twenty five, according to estimates. Musk lawsuit is likely
an attempt to interfere with this funding or to destabilize
open Ai. It's week its point, it's flimsy status as
a nonprofit that will require a great deal of legal
effort to unwind if it's even possible at all, And
(09:36):
I must be clear it may not be possible. I
don't think there's any precedent of anyone ever taking a
nonprofit of this size, of this weirdness connected to like
twenty different for profit entities, and turning it into a
for profit. It just doesn't make sense. But I will
tell you something that might make you happy or might
just make your laugh, which is open AI only has
(09:59):
a a year and a half to do so, a
year and a half to turn from a nonprofit into
a for profit because a year and a half all
that equity they raised, but it's not equity, it's some
weird for profit sharing. Nevertheless, all the money they've raised
in the last round, the six point something billion dollar one, yeah,
it all turns into debt. Oh well, I'm sure they'll
(10:21):
work it out. They sure haven't yet
Aw Zone Media. Hello On, Welcome to this week's Better
Offline Monologue. I'm your host ed ZiT Trump, and I
know some of you are going to say, Ed, didn't
(00:22):
you say we'd get a second part? Didn't you say
we get a second part? Ed? Where's the second part?
It's coming tomorrow. You get a monologue as well. Good Lord.
The complaints from some of you just kidding. You're all
very nice. Now, because I deeply hate myself, I decided
to sit down and read case four to two four
CV zero four seven two to two ygr from the
United States District Court of West the Northern District to California. Nevertheless,
(00:46):
what I'm talking about, of course, is Elon Musk's lawsuit
against Open a I, filed in August. November of last year,
there was an amended complaint. Nevertheless, Elon Musk is alleging
multiple kinds of fraud, as well as violations of the
Sherman Act and to monopoly law from the late eighteen hundreds,
which most notably was the same law that might lead
to the breakup of Google's ad taken search businesses. Elon Musk,
(01:08):
he's suing, and he loves to sue, but in Layman terms.
Musk alleges that Sam Wortman tricked him into funding open
ai as a charity when he actually wanted it to
be more like a for profit entity, a classical start
up model. Musk also alleges a conspiracy by open ai
to stop people who invested in open ai from investing
in other generative AI companies, specifically Musk zons Ai as
(01:30):
long as others like Anthropic. This sounds like some Musky
and bullshit, but this is actually true. It was reported
by The Information and other outlets. The lawsuit itself is contrived,
including annoying things like Musk's lawyers referring to open AI's
tax exempt nonprofit as a for profit market paralyzing Gorgan
Just you don't need to write like this. He fucking
(01:51):
loses anyway. It is pretty interesting, though, and it explores
the deeply weird beginnings of open ai itself. To explain
Joe and was originally founded in twenty fifteen by Elon Musk,
Sam Mortman and a selection of other engineers specifically as
a non profit making open source artificial intelligence, and it
was meant to be a research house. Now another thing
(02:13):
is it was specifically made as a reaction to Google's
acquisition of artificial intelligence firm deep Mind. The plan, according
to emails shared as part of the lawsuit, was to
beat Google to the punch by making artificial general intelligence,
you know, the entirely fictional concept of a conscious autonomous computer,
and then they go and open source it in what
Sam Moortman called an AI Manhattan project. Just I could
(02:36):
go into the history there, but is that really what
you want to compare this anyway? Anyway? Aortmand would go
on to tell Elon Musk that the mission would be
to create the first general AI AGI and use it
for an individual empowerment i e. The distributed version of
the future that seems the safest. More generally, safety should
be a first class requirement. And that is a quote,
(02:58):
by the way, with the technol owned by the foundation
referring to open AI and use for the gift of
the world. Just a lot of bollocks really, Anyway. Things
began to get tense in September twenty sixteen when Sam
Wortman arranged a deal with Microsoft to buy sixty million
dollars of Compute for Well for ten million dollars in
exchange for evangelizing Microsoft as zero as their preferred cloud provider,
(03:21):
along with some sort of vague consultancy services over Microsoft's models.
Musk would respond to the terms by saying, fine with me.
If they don't use the active messaging would be worth
way more than fifty million. Nots seemed like a Microsoft's
marketing bitch. Two months later, Microsoft were put out a
blog post saying that open ai was choosing as zero
as their primary cloud platform and that open ai would
(03:43):
become an early adopter of as zure En series virtual machines,
some of Microsoft's early GPU compute instances. It's been going
quite a while. A year later, in an exhibit from
the trial from September twenty to twenty seventeen, things would
get a little more fraid with Iliasuitskava, a gifted engineer
recruited by Musk in AI's earliest days, sending an email
(04:03):
to both Musk and Aortman sharing concerns about the future.
Aortman worried about how much money would cost the fund.
Open Ai had been considering finding a way to make
it a wouldn't you guess it? For profit entity, but
Sitzkev had other problems and was far more worried about
Altman and Musk. In the email, Suitzkeave of raised concerns
that Elon Musk wanted unilateral absolute control over the AGI,
(04:27):
and that while Musk had claimed otherwise in negotiating how
to keep open Ai going, it was very clear that
and I quote absolute control was extremely important to him.
As an example, Suitzkeaver added that Musk had said that
he needed to be CEO of the new company so
that everyone would know that he was the one in charge,
even though he also stated that he hated being CEO
and would much rather not be CEO. Suitzkeva added the
(04:51):
Musk's concerns that there would be an AGI dictatorship run
by Demis Hassabis, CEO of Deep Mind, but that in
the current structure that Musk was suggest he would become
a dictator if he chose to all very good stuff.
Siitzkav bizarrely then immediately moved on to say something very
very similar to Sam Altman, saying that and I quote
he didn't understand why the CEO title was so important,
(05:14):
and that Sam Altman's reasons had changed, and that it
was really hard to understand what was driving them Siitzkev
also added a question and I quote, is AGI truly
your primary motivation? How does it connect to your political goals?
How has your thought process changed over time? Ortmand would
reassure Musk, both personally and through others, that he remained
(05:35):
focused on open AI's nonprofit mission. In January twenty eighteen,
Orton would suggest a ridiculous idea selling cryptocurrency to fund
open Ai, which Musk would warn would simply result in
a massive loss of a credibility for open Ai everyone
associated with the ico, referring of course, to an initial
coin offering, a flimsy idea that just means just buy
(05:55):
a bunch of tokens before the thing goes live, basically
how crypto works. I guess it was a whole boom.
I'm not doing a fucking podcast about it. Let's move on.
Musk would step down from open ai in February twenty eighteen,
and a month later, Sam Wonan would propose a fixed
maximum term equity race, essentially selling stock in open ai
but an associated entity. Yet it was still a nonprofit
(06:20):
at the time, and that had a maximum amount you
could make on buying it. It's just very confusing and
What it basically means is it means that they would
create an entity on the side that you could raise
money for that would also own all the bits. I'll
get to that in a second. Nevertheless, this is all
extremely dodgy and weird. Around a year later, in twenty nineteen,
(06:43):
Sam Wonman would eventually create the legally precarious for profit
arm of open Ai what I was just talking about,
and it was called open AILP and immediately, according to
Elon Musk's lawsuit, transferred most of the company's assets and stuff.
The same year, open Ai would strike an exclusive partnership
with Microsoft to provide the compute for their models. As
part of the deal, open Ai would give Microsoft full
(07:05):
license to use their pre AGI intellectual property and research,
which is to say, literally everything they've ever made, and
this would in turn make well, this is the funny,
weird part. This is the really crazy. This is the
part that really gets me. They would own everything. Microsoft
would own everything until they hit AGI. Now, AGI at
(07:26):
this point has been defined by open Ai and Microsoft
as when they hit one hundred billion dollars in profit.
Every time I read about and talk about this stuff,
I just think, who is the idiot? Here, is it
Satching the Dellar, Is it Sam Altman or they both
just the kind of mediocre rich guy who just bounces
their skulls together and they say who has the shittiest idea?
(07:47):
Who will be the dumbest boy today. Nevertheless, Microsoft owns
everything open Ai makes until they invent Agi, by which
I mean they make one hundred billion dollars in profit.
It's all so goddamn stupid. It's also stupid. Now at
some point I want to do an entire episode on
this lawsuit because it's got so many exhibits and so
(08:08):
many warring incentives. Elon Musk's Xai competes directly with open
ai to make large language models that no one really
needs and that cost more to run than they will
ever make. And this lawsuit, as with others, features broad
demands for discovering depositions of people at LinkedIn co founder
and former open ai board member Read Hoffman, and attempts
(08:29):
to name both Microsoft and Hoffman himself as co defendants.
Since filing the lawsuit, and the Elon Musk led consortium
of Bias has offered ninety seven point four billion dollars
for the assets of open AI's charity, an often that
would require multiple different government agencies to approve, which open
Aies board has now declined either way. While Musk is
regularly full of ship, he's right about one thing. Sam
(08:52):
Woman clearly had no intention of ever keeping open ai
as nonprofit, nor was he ever dedicated to doing so,
or really anything of them make in himself CEO and
getting a billion dollars. Since twenty nineteen, open ai has
raised over twenty billion dollars in funding, and it's reportedly
raising as much as forty billion dollars in the next round,
led by fucking Masayoshi son of soft Bank. It's so good.
(09:14):
I love it. And they're likely doing this because the
company burned five billion dollars in twenty twenty four, and
it is said to as much as double that in
twenty twenty five, according to estimates. Musk lawsuit is likely
an attempt to interfere with this funding or to destabilize
open Ai. It's week its point, it's flimsy status as
a nonprofit that will require a great deal of legal
effort to unwind if it's even possible at all, And
(09:36):
I must be clear it may not be possible. I
don't think there's any precedent of anyone ever taking a
nonprofit of this size, of this weirdness connected to like
twenty different for profit entities, and turning it into a
for profit. It just doesn't make sense. But I will
tell you something that might make you happy or might
just make your laugh, which is open AI only has
(09:59):
a a year and a half to do so, a
year and a half to turn from a nonprofit into
a for profit because a year and a half all
that equity they raised, but it's not equity, it's some
weird for profit sharing. Nevertheless, all the money they've raised
in the last round, the six point something billion dollar one, yeah,
it all turns into debt. Oh well, I'm sure they'll
(10:21):
work it out. They sure haven't yet
Host
Ed Zitron
Popular Podcasts
United States of Kennedy
United States of Kennedy is a podcast about our cultural fascination with the Kennedy dynasty. Every week, hosts Lyra Smith and George Civeris go into one aspect of the Kennedy story.
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com