Why Disney vs. MidJourney Could Rewrite Hollywood’s IP Playbook

Technically Creative by KoobrikLabs Podcast Transcript

Title: Why Disney vs. MidJourney Could Rewrite Hollywood’s IP Playbook

Orlando Wood (00:00)
My guest today is Rob Rosenberg, a partner at Mozes and Singer and
managing director of MS Strategic Solutions. Rob’s career has spanned
advertising, entertainment, and law, including more than two decades
at Showtime where he rose to be executive vice president and general
counsel while also leading digital strategy. It’s a rare mix. Few
people have had such a front row seat to the evolution of
entertainment, from the rise of streaming to today’s battles over AI,
copyright, and IP.
But what I love about Rob is that he doesn’t just think like a lawyer,
he also thinks like a content creator and a studio head. Rob has
become one of the most thoughtful voices on how IP owners can navigate
this new landscape, advocating for fairness, compensation, and
guardrails while recognizing the opportunities the AI presents at the
same time. his insights in his weekly substack, the Technotainment
Scorecard. And in our conversation, Rob breaks down what the Disney
versus MinJourney case really means.
why asking for forgiveness, not permission won’t fly this time, and
how the deals being made right now could shape the future of
creativity itself. If you want a real deep dive into the legal
challenges being faced by studios and AI companies right now, please
stick around. think you’ll learn a
Orlando (01:11)
Rob, I’m so appreciative that you’re willing to do this, because I’ve
really been wanting to speak to somebody who’s a legal expert on
everything that’s going on. And just, you know, thank you so much for
being here.
Rob Rosenberg (01:21)
I’m happy to be here, thanks for having me.
Orlando (01:23)
Of course. Well, you’re a partner at Moses Singer Limited and managing
director of MS Strategic Solutions. And really what I wanted to kind
of talk to you about is that obviously, you know, there’s a lot of
thought leadership online about, know, what what we’re dealing with
here from a sort of copyright and IP standpoint. And what I think is
so fascinating about you is a little bit similar to myself. I started
in advertising. I then worked in entertainment.
Rob Rosenberg (01:35)
Mm-mm.
Orlando (01:47)
I’ve been back and forth across that line quite a few times. And your
background is very similar and culminated in the entertainment space
and you being at Showtime for a very long time. But now you’re, and
you’ve always been a lawyer, but now you’re really becoming big voice
for all of the copyright realities that we’re facing here, copyright
and IP realities with what we’re facing with AI.

I’ve been back and forth across that line quite a few times. And your
background is very similar and culminated in the entertainment space
and you being at Showtime for a very long time. But now you’re, and
you’ve always been a lawyer, but now you’re really becoming big voice
for all of the copyright realities that we’re facing here, copyright
and IP realities with what we’re facing with AI.
Can you just give us an overview of your kind of perspective on
things?
Rob Rosenberg (02:17)
Sure. So my background, I have been in the content creation business
almost my entire career. I started in advertising like you. I went
into law. I attended law school at night, went to a law firm, then
went to Showtime where I worked in various legal and business roles
for a 22 year period, which was extraordinary.
⁓ eventually becoming EVP and general counsel. During that time,
before I became general counsel, I actually wore two hats at Showtime
where I was SVP of digital strategy as well as being SVP and deputy
general counsel. So, you know, one half of me was business leaning,
one half of me was legal leaning. And, you know, certainly during that
time period was a period when so much change was happening to our
business and
just with devices and platforms and how people actually consume media.
So, you know, I will say that, my slant is definitely from a content
creation perspective. I probably should say that my views are my own
and not those of my firm or my clients or any of that. But, you know,
I’m very much interested in following the cases, the practices, the
deployment of AI through
the creative industries, right?
Orlando (03:38)
Yeah, in your career, I mean, from Ogilvy and Mather to Showtime, how
have those experiences shaped your worldview on the legal challenges
in entertainment technology and AI?
Rob Rosenberg (03:47)
I think that those who create content tend to be very respective,
respectful of copyrights. That they understand that if people are
spending the time, the effort, the artistry in creating content, then
they need to be able to be compensated for all of that effort.
Otherwise, people will all become lawyers and all become, you know,
stockbrokers. So if…
If we make it more and more difficult for artists to receive the
financial benefits of their efforts, then we’re basically doing the
creative industries a huge disservice. And considering it’s one of the
industries that the US leads the world in, it would really be foolish
for us to take for granted something that has really been a leading
industry for the United States.

Orlando (04:39)
Yeah, and I think that we’re dealing with two separate issues,
separate but component issues, aren’t we? There’s obviously the
remuneration aspect of the credit, the remuneration. And I think that
the ingestion maybe is potentially fair use, but at the end of the
day, the learnings from what they’ve ingested, they are now monetizing
and they’re not giving credit.
Rob Rosenberg (05:00)
Mm-hmm.
Orlando (05:02)
consent or compensation to any of the artists that they’ve taken this
work from. How do you unpack that? Because it’s sort something very
clear happened, a technology that we can’t possibly comprehend lives
in the middle, and then something that we understand was wrong
happened on the other end. But this question mark in the middle covers
a lot of sins, doesn’t it?
Rob Rosenberg (05:24)
It 100 % does. think if you wanted to use a piece of music in your
podcast, you would go out and you would license that piece of music.
If you wanted to create a new piece of content or a new product and it
included music or it included clips of video, know, unless it was
specifically for one of the fair use exceptions to copyright, you
would go out and you would license those clips to use them. You’d sell
advertising.
on your podcast, whatever it is. And I think what happened is that the
AI platforms began to train these models on content that was, in some
cases it was licensed, but in many cases it was scraped from the
internet, it was purchased from pirate sites. So they built this new
industry, this new product using people’s copyrighted works in many
cases without
obtaining proper licenses. And now that they’ve done it, now that
they’ve launched it, now that there are hundreds of millions of users
every day of these products, saying, well, how could we go back and
just, how can we start over from scratch with this? can’t. And
therefore, it’s completely impractical. And therefore, you shouldn’t
hold us accountable for this, which
To me, it’s kind of like the automotive industry saying, hey, we can
make really cheap cars if we don’t have to pay for the parts, right?
We really want the US to be a leader in auto manufacturing, so why
don’t you just make it a zero cost part for all the US manufacturers?
It will be golden. We’ll have a great business here.
Orlando (06:50)

Right. ⁓
Yeah.
Yeah.
Rob Rosenberg (07:07)
You know, unfortunately that’s not the way the world works. And I
think we have seen other industries like the music industry with
statutory licenses, you you can set up a scheme where the AI platforms
can, you know, certainly, you know, working backwards, they can
compensate people whose works that they used, but, you know, and then
on a going forward basis, they should be giving people the option.
of whether they want their works included or not. And I think part of
that also is the notion of what are they outputting to consumers from
the platforms they train. But we can handle that as a separate
question.
Orlando (07:48)
Yes, exactly. Yeah, I mean, it’s funny because there’s that old adage
of asking for forgiveness rather than permission, but it doesn’t even
seem like they’re asking for forgiveness. They’ve skipped to just
expecting that this is totally OK what they’ve done.
Rob Rosenberg (07:58)
Absolutely.
Yeah. And, you know, to a certain extent, as you’re aware, like there
have been a couple recent decisions in the Northern District of
California that have come down on that side. And they said, look,
taking these books and using these books to train these large language
models, they were transformative. were not, know, you’re not going to
be able to replicate the books. Like I think in one of the cases,
there was even a ⁓ filter that was put in place where
in the case of something that the material that the platform was
trained on, that you couldn’t reproduce more than 50 words from the
original source. So it seemed like they recognized the fact that for
this to be legal and transformative, then that it really needed to be
something that was ingested and used to produce something new and not
potentially impact the
the market for the original works.
Orlando (08:58)
Yeah, I mean, it’s funny because I think that one of the things that
they’re also doing is they’re sort of claiming that, it was nominal,
know, what we put in there, you know, it had some impact, but really
we had developed the technology and then this creative stuff was sort
of draped on top to kind of make it nice. But I think that they
realized they could have had a very good model trained solely on
publicly available legal texts and, you know, and…

Yeah, I mean, it’s funny because I think that one of the things that
they’re also doing is they’re sort of claiming that, it was nominal,
know, what we put in there, you know, it had some impact, but really
we had developed the technology and then this creative stuff was sort
of draped on top to kind of make it nice. But I think that they
realized they could have had a very good model trained solely on
publicly available legal texts and, you know, and…
Rob Rosenberg (09:23)
Hmm.
Orlando (09:25)
City Hall filings and all of those sorts of things that are in the
public domain. But I think they realized that the stuff that would
give it life, the stuff that would get people hooked was the stuff of
life. It was poetry and literature and art and the things that we seek
out and pine for and want to be a part of. And I think they realized
that. And yet now they’re trying to claim that this is…
Rob Rosenberg (09:38)
100%.
Orlando (09:49)
was just, well, it was the last 1%. But I think in that 1 % is sort of
everything.
Rob Rosenberg (09:54)
Yeah, I agree with you that I think it’s the kinds of things that make
this sexy, that make this attractive, that make people want to dip
their toe in and try it out. If this were just a better search engine,
then you might get some people to try it, you might not get people to
try it. But the fact that they trained it on all, you know,
Orlando (10:01)
Mm. Yeah.
Rob Rosenberg (10:17)
the world’s libraries, you know, it was in the entire internet. think
it’s those things that made people intrigued enough to dip in, you
know, and, you know, let’s not forget, they aren’t doing this out of
the kindness of their hearts, you know, like these are, you know,
billion dollar companies now, you know, like they built this thing,
like, you know, their valuations are ridiculous and the hundreds of
billions of dollars they can certainly afford.
to compensate people whose works they’ve used.
Orlando (10:47)
Yeah. mean, but isn’t part of the problem that, because you mentioned,
you know, obviously taking a piece of content, training on it and then
outputting it. And a lot of people kind of say, well, but isn’t that
the entire creative process for a human? know, mean, Quentin
Tarantino, you can go online and see super cuts of shots. He’s
effectively sort of, you know, stolen or changed, but it’s, you know,
there’s, there’s that old adage that, you know, great writers borrow
fantastic writers steal outright.

Yeah. mean, but isn’t part of the problem that, because you mentioned,
you know, obviously taking a piece of content, training on it and then
outputting it. And a lot of people kind of say, well, but isn’t that
the entire creative process for a human? know, mean, Quentin
Tarantino, you can go online and see super cuts of shots. He’s
effectively sort of, you know, stolen or changed, but it’s, you know,
there’s, there’s that old adage that, you know, great writers borrow
fantastic writers steal outright.
You know, and there is that love there, there is that homage, but you
know, that process of being inspired by something and then outputting
something that is demonstrably similar, doesn’t that highlight the
dual purpose of copyright law? On the one hand, copyright law is
supposed to protect people who have created things, but it’s also
meant to be fuzzy at the edges. It’s meant to inspire further
creativity. It’s meant to allow you to build on what’s come before.
Rob Rosenberg (11:33)
Mm-hmm.
Orlando (11:42)
And how does one deal with that dual nature in copyright law, it being
a servant of those two masters?
Rob Rosenberg (11:52)
Sure, there are, you know, there’s the recent Disney and Universal
case against ⁓ mid-journey. And I think that’s really gonna take this
legal battle to the next level. I think that, you know, I jokingly say
we’re now at, show me the output space of this analysis where it’s one
thing to say what goes in, fine. You know, you can put a ton of stuff
into these models
Orlando (11:58)
mid-journey.
Rob Rosenberg (12:18)
And again, I’m not conceding that that’s the final word. You we’ve
heard from two cases from a court that is in Silicon Valley. It’ll be
very interesting to see if there are other courts. Yeah, yeah, yeah,
that’s the meta case. So there are two cases there. And I think it’ll
be interesting to see if other courts in other jurisdictions
Orlando (12:29)
That’s the anthropic case, is that right? Yeah.
Rob Rosenberg (12:43)
land the same way that they do. But again, both of those cases really
dealt with the input. They dealt with the content that was used to
train the model. And what I think the Disney and Universal case is
doing, and I think it’s going to be very important and may actually
change the analysis, is it’s really focused on the outputs. It’s
focused on what are these models outputting to consumers? And
You know, in the case of visual arts, you know, their case says, you
you, you fed Mickey Mouse, you fed Darth Vader, you fed the Disney
princesses into your model. And now you’re spitting out outputs that,
know, whether you call it them or not, everybody’s going to recognize
Spider-Man. Everybody’s going to recognize the minions, you know,
that, that you’re, basically producing outputs for people. You’re
making money. You’re charging people.

You know, in the case of visual arts, you know, their case says, you
you, you fed Mickey Mouse, you fed Darth Vader, you fed the Disney
princesses into your model. And now you’re spitting out outputs that,
know, whether you call it them or not, everybody’s going to recognize
Spider-Man. Everybody’s going to recognize the minions, you know,
that, that you’re, basically producing outputs for people. You’re
making money. You’re charging people.
Orlando (13:23)
Mm-hmm.
Rob Rosenberg (13:38)
to use your product and you don’t have any rights to create derivative
works of my protected IP.
Orlando (13:45)
Yeah, and it doesn’t really matter if you did that intentionally or
not. If you’ve put an internet scraper in there, if you search for
Darth Vader, an archetype will come up. That archetype was created by
Disney. It sort of doesn’t matter if you’ve ingested somebody’s
drawing that they put on their blog. If it is demonstrably similar to
what Disney has created in that archetype and that look, it is, well,
Lucasfilm now a Disney subsidiary.
Rob Rosenberg (13:52)
Mm-hmm.
Hmm.
Yeah, sure.
Orlando (14:13)
You know, it sort of doesn’t matter, you know, whether the training
set came from Disney or not or ripped off Disney directly or not.
Rob Rosenberg (14:22)
Right. I’m 100 % in agreement that I think, you know, one of the
factors that courts look at when they’re determining whether something
is fair use, you know, yes, these first two cases really hung their
hats on the fact that was this use transformative? Did they take these
books, input them into the models, and you’re really not seeing
anybody spit out a copy of a book that they otherwise would have to
buy? But
Orlando (14:48)
Mmm.
Rob Rosenberg (14:48)
But one of the fair use factors is the impact on the market for the
original works. And I think with this new case from Disney and
Universal, with other cases that have been presented, that what we’re
seeing is a big focus on, you producing things to consumers that you
are charging consumers for that I could charge people for?

It shouldn’t be something that just because it was okay for you to
train on it, that you could produce an output, an end product that
consumers take away as satisfied customers of your service without
there being any proper licensing or compensation here.
Orlando (15:30)
Right. And I mean, the choice of Mid Journey, is that just because
that was a big brand that they could go after that couldn’t possibly
defend itself? know, I mean, obviously it can, but it can’t on the
same level as Google Gemini, which VO3 is obviously outputting a
similar level of quality, but VO3 is backed by Google and Disney going
toe to toe with Google is going to be a trickier
Rob Rosenberg (15:33)
.
Orlando (15:57)
you know, a trickier fight than it is with Mid Journey. Do you think
that’s the play here?
Rob Rosenberg (16:01)
Yeah, you know, I don’t represent Disney or Universal. So, you know, I
haven’t, I wasn’t in the room when, they were determining their
strategy. I do think that Mid Journey is, at least has been an early
provider of visual output that the either, either images or video. And
I think because of that, there were a lot of examples of the kind of
harms that Disney and Universal
Orlando (16:04)
Right now.
Rob Rosenberg (16:28)
were feeling from this platform and these types of platforms that they
felt that they could make a very strong case going in. And certainly,
I don’t believe that what, even though Mid Journey may be smaller than
Google or maybe smaller than OpenAI, but I don’t believe that there’s
any difference in terms of the analysis of the courts, the analysis of
the law.
So it shouldn’t make a difference whether you attacked one or the
other. It just may be that they were able to make the strongest case
at this point in time.
Orlando (17:02)
right. That was very deftly handled. I think, you know, in everybody
else’s mind, it’s like, well, the bigger one’s gonna win. I think. But
yes, no, you’re absolutely right. From a legal and argumentative
standpoint, there is no there is no and should be no difference
between the two. But, you know, I mean, in terms of this, you know, so
obviously, there’s going to be a gray area until these things are
resolved. And like you said,

right. That was very deftly handled. I think, you know, in everybody
else’s mind, it’s like, well, the bigger one’s gonna win. I think. But
yes, no, you’re absolutely right. From a legal and argumentative
standpoint, there is no there is no and should be no difference
between the two. But, you know, I mean, in terms of this, you know, so
obviously, there’s going to be a gray area until these things are
resolved. And like you said,
Rob Rosenberg (17:05)
Thank you.
Mm-hmm.
Sure.
Orlando (17:26)
Anthropic just had a win in its favor and met in its favor from a
Silicon Valley court saying that their ingestion was fair use. didn’t
deal with the generation side so much. Now we’re waiting on the Disney
and NBC Universal versus Mid Journey. In terms of best practices in
this gray area period, mean, we’re certainly speaking to lots of
different studios, creators, most of them
are taking what I think is the right approach, which is they are sort
of doing nothing on an output standpoint. They’re privately looking at
ways to use these tools to improve their workflows. They’re looking at
ways to engage with these tools so that they stay up to date on what
the capabilities are. But they certainly all have a view of nothing
goes out the door that is generative AI or
certainly generative AI that might have the original sin of copyright
theft at its heart. Is that the only way to be going about things
right now?
Rob Rosenberg (18:25)
I think so. have been, my colleagues in the legal profession aren’t
going to like this, but the only people who win in litigation
consistently are lawyers. All of this uncertainty, all of this
litigation that is happening right now, it’s not great for business.
The AI ⁓ platform, the AI platforms would love for
Hollywood to embrace them and to engage with them and to, you know,
fully, you know, utilize them to the extent of their capabilities. But
I think because of all the litigation, because of all the uncertainty,
it’s really put them into a, you know, all right, you know, we’re
going to, we’re going to experiment. We’re going to, we’re going to
dance around the outside of this to make sure we understand
everything. But we’re really not fully on board with people who just
think that
You know, they can take, ⁓ you know, the jewels of our crown and make
a business off of them without participating with us.
Orlando (19:28)
Yeah, yeah. I mean, so, you know, are there emerging legal standards
or best practices that, you know, I mean, is there, you know, I think
that a lot of these things are being talked about and I think there
are various levels of creating, of being created, but like, is there a
copyright tracking similarity function, you know, system, you know,
that

Yeah, yeah. I mean, so, you know, are there emerging legal standards
or best practices that, you know, I mean, is there, you know, I think
that a lot of these things are being talked about and I think there
are various levels of creating, of being created, but like, is there a
copyright tracking similarity function, you know, system, you know,
that
that can even be used at the moment. mean, it’s kind of, do you see
any frameworks emerging that might give us a glimpse into what the
future might look like in terms of compensation, attribution, and
consent?
Rob Rosenberg (20:05)
Yeah, I mean, I think the best path forward is let’s make a deal. You
know, like, let’s get the sides together. Let’s figure this out. Let’s
come to terms. There is certainly money to go around and certainty is
better for everyone. You know, it will be better for the creative
communities. It will be better for the AI platforms that, you know,
all of this guessing and waiting.
for the courts to decide things is really put a big stall, a big drag
on the deployment of these technologies. So I do think that there’s an
opportunity for people to come to the table. We’ve already seen
reports where there are companies that are licensing, that when the
platforms are willing to step up and take licenses, then they’re
dealing with the issue of compensation. They’re dealing with the
issues of credit.
And they’re also like in these agreements, when we’re talking to
clients, we’re talking to them about these agreements should include
provisions around what are the platforms going to output, what are
they allowed to, what are they not allowed to do, and what are the
parameters around those. So you’re putting some guardrails around your
core business while also giving them the certainty that they can move
forward with your content.
Orlando (21:25)
Mm-hmm. I mean, let’s say for a moment that Disney and NBCUniversal
wins. The compensation waterfalls for most of their large pieces of IP
are unique down to the platform. know, I mean, it’s very possible that
a single producer might have the option or rights to exploit a
superhero, for example, on film. Now, VO3 is outputting not…
Rob Rosenberg (21:48)
Mm-hmm.
Orlando (21:50)
traditional box office and theatrically released films, but it is a
moving image in a digital file. know, presumably that, and then that
producer also has a responsibility of further remuneration to the
artists themselves potentially, or underlying rights holders. I mean,
how do you even begin to work out those sorts of waterfalls in a,

traditional box office and theatrically released films, but it is a
moving image in a digital file. know, presumably that, and then that
producer also has a responsibility of further remuneration to the
artists themselves potentially, or underlying rights holders. I mean,
how do you even begin to work out those sorts of waterfalls in a,
In a deal like this, how can you, in terms of the damages done, know,
were Disney to win, how does it even go about trickling that back? Or
is it just creating a framework that allows for, you know, that
waterfall to be siphoned off in the right way in the future? And it’s,
anytime you output, yeah, sorry.
Rob Rosenberg (22:32)
Yeah, I
know that’s okay. I’m sorry. I didn’t mean to cut you off, but yeah, I
think that we are going to see there’ll be some sort of result. Let’s
say that Disney and Universal win, you know, there will be some
payments required from, you know, from mid journey. And then on a
going forward basis, it will also give them the standing and the
fortitude to say to the mid journeys and all of the other
Orlando (22:36)
No, no, no.
Rob Rosenberg (22:58)
platforms that are outputting images, video, et cetera, to say, look,
I license this stuff all the time, all day long. Disney’s licensing
program is, I think, one of the biggest in the world. So I’m happy to
get into the licensing business with you. You want to let your users
create fan fiction shorts with superheroes and Star Wars characters?
Who wouldn’t want that? It will bring.
so many people to your platform, it will give consumers who are
already fans of a particular IP a place to aggregate and show their
love for the IP. So there’s a lot to be won here by them creating this
scheme around how do they license that and move forward from it.
Orlando (23:45)
Yeah, and I think as well create guardrails around it. I mean, at the
end of the day, you you have to be making sure that people aren’t
doing things with your IP that are sort of unspeakable, lewd. mean,
you know, things like that. These are, know, you can’t have it
denigrate the IP and all licensing deals include that. Yeah.
Rob Rosenberg (23:49)
100%.
Yes, they do.
I think we’ve already seen some of that. And I think even some of that
is mentioned in the Disney Universal case.
Orlando (24:11)

Mm-hmm,
yeah. Gosh, that, mean, to be honest, your answer just then took out
like three of my key questions over here. So yeah, you’re doing
incredibly well in terms of, in the shooting gallery of ticking off my
questions. ⁓ But yeah, I mean, so.
Rob Rosenberg (24:18)
Yeah.
All right.
Orlando (24:29)
So what are the most significant misuses do you think at the moment? I
do you think everybody’s equally culpable or do you look at the field
and go, these people are particularly bad or there are actions that
some are taking that are aggressively against fair use. I think that
actually, what’s your feeling about that?
Rob Rosenberg (24:49)
It’s a difficult question to answer because these platforms and these
models are so opaque into how do they actually work? know, like there
are certainly different companies are applying different filters
inside of them. So I think, you maybe you ask the same query of, you
know, three or four or five AI platforms and you see, they all
producing the same result or are…
some more onerous or some more respectful of IP. think that, you know,
I personally haven’t done that analysis because also the models are
constantly evolving and changing. So, you know, what happens today may
be very different when version 4.0 comes out tomorrow, you know? So I
think that that’s a difficult question to answer at the moment. think,
you know, we haven’t talked about the fact that, you
Deep fakes are a huge problem for the artist community, for public
figures, even for ordinary citizens. There have already been instances
cited all over the place where somebody gets into a beef with the
school principal and all of a sudden there’s a video or a sound
recording that comes out that they did something reprehensible and it
turned out that it was completely fake.
Orlando (26:00)
Mm-hmm.
Rob Rosenberg (26:08)
You know, I think that there needs to be protections built in to the
system to deal with things like that as well.
Orlando (26:15)
Right. mean, are there corporate governments, I mean, given your work,
you know, and what you do every day, are there corporate governance
models or policies you recommend organizations adopt to mitigate a
risk around AI usage?

Right. mean, are there corporate governments, I mean, given your work,
you know, and what you do every day, are there corporate governance
models or policies you recommend organizations adopt to mitigate a
risk around AI usage?
Rob Rosenberg (26:31)
Yes, what I would say is, again, I think that the people who are most
respectful of copyright are people who are in the copyright business.
That I think that they recognize and they won’t do something with your
copyrighted material because they know very quickly that says, you
know, the gates are open and anybody can do that with my copyrighted
material as well. So,
I think that there is a very strong desire to, mean, people, AI isn’t
going away. It’s not a fad, it’s not going away. It’s gonna impact
every business that you can think of. So I think people recognize that
and all they’re trying to do is say, let’s do this fairly responsibly,
safely. And I think those are the kinds of corporate governance
checklists that
Orlando (27:00)
Yeah.
Rob Rosenberg (27:22)
we’re advocating to companies both on the development side and on the
content side.
Orlando (27:29)
Yeah, I mean, I think, you know, it’s funny because you’re you
obviously sort of, you know, we’re we’re at showtime at a sort of key
moment where you got a legal strategy through the rise of streaming
and OTT platforms and things like that. What was interesting to me is
we were at a studio where somebody one of the model, the the AI
companies had approached them and said they wanted to license their
stuff for training. And they said, we’ve just been through this like
we licensed all of our stuff to Netflix.
Rob Rosenberg (27:57)
hehe
Orlando (27:57)
and how
Netflix is eating our lunch. know, like we’re a little bit once
bitten, twice shy, you know? And I do think that there’s this degree
to which we’ve just seen one transformative shift in the move to
streaming and then, know, you know, direct distribution platforms. And
now we’re seeing a similar, you know, playbook play out with the AI
models. And I was quite encouraged that they said, listen, we’re not
gonna do it again.

Rob Rosenberg (28:00)
Mm-hmm.
Orlando (28:22)
We’re not gonna give you all of our news and journalism work just so
that you can eat our lunch again. And I think it was encouraging that
they had learned from that mistake. I mean, what’s your, you obviously
live through that big shift. What do you think, are there any other
learnings we can take from?
Rob Rosenberg (28:22)
Mm-hmm.
Mm-hmm.
I think we’re going through some of that, but I also think it’s
somewhat different because in the case of Netflix and people couldn’t
say no to the money. All of a sudden, most content is licensed by
Windows, so sequential Windows, they’re licensed by platforms.
And all of a sudden there was something new here that represented a
new opportunity for studios, for networks to find another window,
another source of revenue from content that was already in the can. It
didn’t cost them a dollar more to license that stuff to Netflix. Of
course, what happened is Netflix built
this rabid following of consumers because they offered a lot of
content at a low value and offer, I’m sorry, at a low cost, which
represented a tremendous value to consumers. They said, for $7.99,
look at all the stuff I get. And I think because of that, and at the
time, let’s remember that when Wall Street was evaluating Netflix,
Orlando (29:41)
Yeah.
Rob Rosenberg (29:49)
It was really looking, is Netflix still growing subscribers? know,
from quarter to quarter, that was really the benchmark for them. And
it wasn’t really looking at, are they making more money? It was
really, they were evaluated on a different scale than the programmers
were at that moment in time, which enabled them to offer all this
value.
and still keep their price relatively low. So, you know, I think that
that’s very different. You know, here we’re talking about something
entirely new. I don’t think, you know, we’re not talking about
licensing content to an AI platform so people can watch that content.
You know, there are plenty of avenues to do that. You know, maybe some
of them will get into that business as well. But I think it’s really
about

Orlando (30:15)
Yeah.
Rob Rosenberg (30:37)
How can they utilize the content and the technology to empower content
creators, the high schoolers, the film students, the writers, the
artists? How do they use the technology of AI to take their IP a step
further?
Orlando (30:59)
Mm-hmm.
Yeah, I mean, it’s it’s I’m sort of interested. You know, there’s
obviously a lot of focus on on sort of specific artists, you know, and
a single creator. But obviously being in the creative communities, we
know that it takes a village to to create, you know, to create
everything. mean, you know, let’s say a studio trains a model on its
own back catalog, hundreds of hours of content it owns legally clean
shore. But what are the sort of
labor implications, is there any ethical responsibility to the
industry at large to ensure that the broader people who have created
these things, the teams that have created these things are either
compensated or there’s still a place for them somehow?
Rob Rosenberg (31:42)
Yeah, I so I think there’s two things that are in your question. One
is talking about profit participants, the people who have today for
each exploitation of a given work, depending upon what people
contributed to that work, that there are some people who are owed
additional compensation for new exploitations of the work. And in
those cases, I agree with you if there is revenue that’s associated
with
the exploitation of a given work, then they’re gonna have to look long
and hard about what their obligations are, but they would likely have
to compensate those profit participants. I think the second element,
which we hadn’t talked about yet, is this whole fear of, know, is AI
gonna eat everybody’s job? know, like it’s, and I think it’s not a
chicken little story. I think that there is a lot that
You know, like there have been quotes about entry level jobs, not even
talking about Hollywood, but just entry level jobs generally that I
think there was a quote from, I can’t remember, I believe it was
OpenAI, where they were saying that, you know, within five years, all
the entry level jobs will be taken over by AI, you know. So AI is a
fantastic tool. And I think that’s the way people are viewing it right
now.

But if it becomes a tool that all of a sudden is doing work that was
previously done by artists and artisans, I think there is a big
concern. You know, number one is, you know, is there going to be this
huge decline in job opportunities? You know, I think that, you know,
some people are making the argument, well, costs less to make
the big blockbuster movie, we can make more big blockbuster movies. So
there’s more opportunity for people who want to act and direct and art
directors and producers and crew members, et cetera, et cetera. So I
think it’s still too early in the business to understand how it’s
gonna shake out. I think that…
universally there has been a sentiment by the creative community that
they haven’t seen anything yet that makes them think that the empathy,
the creativity, the humanness that artists bring to creative works is
in danger of being taken over by AI.
Orlando (34:13)
Yeah. Yeah, I mean, somebody said something the other day that I
actually, I thought of, and they said, every tool we’ve ever created
helped us do more work. We’ve never, we’ve never as a species
developed a tool and done much less with it. know, so that gave me
some pause. Then again, I think we do spend a lot of time watching
television, you know, so.
Rob Rosenberg (34:23)
Mm-hmm.
Mm-hmm.
Orlando (34:35)
Are you seeing, are you aware of any new clauses popping up in
entertainment contracts to protect against AI misuse? We know of an
agency that represents talent that they actually rework their nudity
rider ⁓ to account for potential AI training use, anything like that.
Are you seeing any trends in that direction?
Rob Rosenberg (34:49)
Mm-hmm.
So obviously when SAG renegotiated their agreements in 2023, there
were new AI provisions that were added to that. And I think we’re
seeing that filtered in contracts that are created today. I think
we’re starting to see some agencies that are offering services to
their clients, to police.
what is out there and what’s being produced with their name image
likeness and to go after that for them. I think that again, the
studios ultimately are looking for flexibility. If there’s tools that
enable them to do certain things, they wanna be able to do it. And I
think that the actor community has said, we’re not necessarily opposed
to that, but…

what is out there and what’s being produced with their name image
likeness and to go after that for them. I think that again, the
studios ultimately are looking for flexibility. If there’s tools that
enable them to do certain things, they wanna be able to do it. And I
think that the actor community has said, we’re not necessarily opposed
to that, but…
We need, you need our permission. You need to compensate us and you
need to give us credit for it.
Orlando (35:52)
Yeah, do you think the unions have gone far enough? mean, you know, we
had the WGA, the SAG strikes that, you know, we’re only what? Nine
months to a year on from now. And it sort of felt like there was a lot
in the press about this commentary, but it felt like the collective
bargaining deals that came out of it were a little bit toothless.
What’s your feeling on that?
Rob Rosenberg (36:13)
Yeah, think they, think the part of the problem is the moment in time
that they were in when these things were getting negotiated because
we, and I still think we’re a little bit premature in understanding
the real implications of AI, but I think that they did ⁓ a great job
in getting what they could get at that moment in time.
to protect their members in those new agreements.
Orlando (36:39)
Yeah. Yeah, because it feels like, you know, the I know James Earl
Jones licensed his voice or his estate license his voice so he can now
always be Darth Vader. know, that presumably will generate revenue for
his estate, either in one big chunk or, you know, in a drip feed over
time. It feels.
Rob Rosenberg (36:51)
Yep.
Orlando (37:00)
You know, on the one hand, feels like we’ve been here before. know,
Schuster and and I’m forgetting his name. Comic nerds are going to be
so upset with me. But Schuster and Segal, who did Superman, you know,
they famously sold Superman for one hundred dollars. Now, granted,
during their time, they were earning what would be now the equivalent
of a million dollars a year. So they were compensated during their
time. And then, you know, they they sued.
Rob Rosenberg (37:11)
See you.
Orlando (37:26)
afterwards. I guess the difficulty is there is this notion that
someday Siegel and Schuster would pass away and some other artists
would have to draw and write Superman. I guess the problem with an AI
artist camping out in a role for all of eternity is that then there is
no stepping stone. There’s no ability for, you know, some new actor to
come in and

afterwards. I guess the difficulty is there is this notion that
someday Siegel and Schuster would pass away and some other artists
would have to draw and write Superman. I guess the problem with an AI
artist camping out in a role for all of eternity is that then there is
no stepping stone. There’s no ability for, you know, some new actor to
come in and
Rob Rosenberg (37:33)
Mm-hmm.
Orlando (37:51)
David Corrin’s sweat might be the last ever Superman. ⁓ And that I
have trouble wrestling with. But then again, maybe we’ve created a
tool that has the ability to make so many new pieces of IP that it
sort of doesn’t matter. We’ll have this infinite new developing
catalog of new franchisable IP.
Rob Rosenberg (37:55)
Mm-hmm.
And at least the current state of the law from the US Copyright Office
is that AI created works are not protected under copyright. There are
some exceptions being made like where there’s enough human input,
human interactivity to it that creates independently protectable
elements of that work. But I do think, like you said, that it’s
You’re saying there won’t be anybody else to write Superman, but in
fact, everybody will be able to write Superman, you know? Like if it’s
not protectable and this is all just, you know, something that is fair
game, then there’s nothing, you know, all of these exclusive rights
that are given to copyright owners are no longer given to somebody.
you know, I think
Orlando (38:44)
Yeah.
Rob Rosenberg (39:03)
IP that is created directly from AI, there’s a big, huge question mark
about what the future means for those.
Orlando (39:11)
Yeah, because I guess in my hypothetical, was that if Siegel and
Schuster, before they died, had the ability to create a GPT that could
write Superman forever, that would be the kind of, you know, that no
one else would get the chance. You know, I mean, I think that also
presumes that we’re not a populace that is easily bored with one point
of view and we will naturally gravitate towards the rolling of the
dice on something new.
Rob Rosenberg (39:22)
Mm-hmm.
Right, right.
Hehe.

Hehe.
Orlando (39:39)
⁓
Yeah, that’s ⁓ fascinating. Yeah, I mean, so let’s say you were
advising a major network something like Showtime. you know, I guess
the advice it seems to me based on one of your questions earlier would
be stay interested, but take no generative action until there is
regulatory clarity either on the federal level or at least on the sort
of the state court level.
you know, and with these big things coming out, what do you think the
timeframe is? I mean, how long are we gonna be in purgatory with
these, with this question?
Rob Rosenberg (40:09)
⁓
I mean, you know, I’m going to answer your question that you answered
differently, and that I’m going to say, like, you know, I’m a deal
guy, you know, like I would say to my clients, there’s a deal to be
had here. You know, like, you could join all of the lawsuits that are
out there, or you can go and find out if there’s an appetite for
somebody to license your IP under terms that are acceptable to you,
you know, like, like,
Orlando (40:16)
Right, okay.
Rob Rosenberg (40:36)
Like frame them out, put in guardrails, make sure that you’re not
gonna be kicking yourself 10 years from now saying, shoot, why did we
do that? But at the same time, there is likely a deal to be had today
with these platforms. And I would say to them, go out and make a deal.
If now, because of these decisions in the Northern District of
California, all of a sudden the AI platforms are emboldened.
And they’re saying, you know, I’m not going to make any more deals for
a while and I’m going to let this play out in the courts. Then so be
it. Then I think then you have to let this go a little longer and see
if there are other courts that are going to take different positions
than the Northern District of California.
Orlando (41:20)
Do you think that that’s maybe why some of them are waiting? So let’s
say, for example, Sony is waiting to use these things. Are they kind

of going, I could make a deal, but I want to see how Disney and Mid-
Journey turns out because if they’re going to write a massive check,

I’ll do the deal after I get the massive check. Do you think that’s
their current strategic posturing?

Do you think that that’s maybe why some of them are waiting? So let’s
say, for example, Sony is waiting to use these things. Are they kind

of going, I could make a deal, but I want to see how Disney and Mid-
Journey turns out because if they’re going to write a massive check,

I’ll do the deal after I get the massive check. Do you think that’s
their current strategic posturing?
Rob Rosenberg (41:37)
I guess, yeah,
I mean, I guess it’s possible. I don’t know that that really is what’s
going on there. think that, you know, certainly until the case was
filed that, you know, I don’t know how much it was known as to, you
know, who was gonna file what cases where over what types of claims.
So I think, you know, you know, the studios have been around, you
know, we’re hope that the studios continue to be around.
You know, this is whether you believe theaters are dying or not dying
or whatever, people still love content, you know, and we’ve seen that
with the rise of YouTube. We’ve seen that with the rise of many
different platforms that people have things to say, people have
things, stories to tell. And I think that they’re going to be around
for a long time. you know, I think it would be a disservice to them to
just say, I’m going to
freeze this moment in time until I let this thing play out because it
could be years before there’s some sort of resolution here. I really
do think that if I were advising my clients today, I would say to
them, let’s see if there’s a deal to be made on terms that you think
you’d be happy with.
Orlando (42:51)
Yeah, because I think that the other thing is, you know, at the end of
the day, the world is spinning slowly on its axis and, you know, time
matters, especially because they’re currently the last paradigm shift.
And we talked about it before with streaming is essentially
distribution became free. These companies, especially the big studios,
are on the one hand content generation shop. But on the other hand,
they were.
incredibly sophisticated distribution platforms, and that’s where a
huge amount of their revenue came from. ⁓ And now distribution has
become essentially free. So it’s sort of cutting off one of the legs
of the stool. I think that they need to be generating things,
generating content faster and cheaper than ever before. And so you’re
right, time is a factor because they need to be…
Rob Rosenberg (43:19)
Mm-hmm.
Orlando (43:40)
seeing these efficiencies. One of the other things that we’ve noticed
is usually when people come to us and our whole res indetra is to help
legacy companies and creative companies that have been doing this a
long time upskill and move into the AI world. But they’re usually
coming to us after they’ve let people go. And that’s just proof that
the model is what’s broken. It’s not their hiring practices. It’s not
their

seeing these efficiencies. One of the other things that we’ve noticed
is usually when people come to us and our whole res indetra is to help
legacy companies and creative companies that have been doing this a
long time upskill and move into the AI world. But they’re usually
coming to us after they’ve let people go. And that’s just proof that
the model is what’s broken. It’s not their hiring practices. It’s not
their
Rob Rosenberg (44:03)
Mm-hmm.
Orlando (44:05)
their money-grubbing, know, budget margin-conscious, you know, bean
counters in the back room that are saying, get rid of those people,
because we could do it with an AI. It’s exactly the other way around.
They’re coming to us going, we’ve had to let people go because
everything’s, you know, the sky is falling and we need help. We need
to do the same amount of work, if not more, with now X amount
reduction in staff.
Rob Rosenberg (44:28)
And I think they’re all in that process. I’m going to go back and just
disagree with you. I’m not ready to concede that distribution is free.
you know, like, you know, I think that as somebody who…
Orlando (44:36)
No, no, yes, I said most most yes, you’re right. I should have said
getting
closer to free than it was.
Rob Rosenberg (44:43)
Yeah,
I think that there’s still a lot of people who are paying for the
rights. You know, they’re properly licensing content to distribute
content through proper channels. And I still think that is a huge
business for the studios, for the networks, et cetera, and will
continue to be so for long time. I think that viewing habits have
changed, are changing. I think that the devices, the time,
that people a lot in their day and the things that are competing for
those hours in the day have changed. And I think in each of these
cases, each of these studios, each of these programmers need to be
pivoting to the new reality. And that’s why I say, you I think for
many of them to just sit around and wait, you know, like you either do
things or they get done to you.
And I think that in a business like this where things are moving
rapidly, you need to make the best decisions strategically that you
can. You need to look into the tea leaves the best that you can. But
waiting around for something to happen is probably not the right
decision long term.

Orlando (45:58)
Yeah. Now, I mean, you know, knowing, knowing somebody of, of your
stature and, and, you know, that you’ve been thinking about this a
lot. you know, I, and also seeing, you know, I do this thing myself
where there’s usually something that I, you know, at, ⁓ at over drinks
or something where you go, I’ll tell you the one thing nobody’s paying
attention to, and this is going to be the thing in X many years.
What’s your one thing that.
people are maybe missing right now that you think is a bigger problem
they’re paying attention to.
Rob Rosenberg (46:30)
⁓
I think there’s an opportunity right now in that a lot of the large
corporations are spinning off cable networks and they’re spinning off,
you know, some of them with those cable networks come content
libraries that are associated with that network, et cetera. You know,
there is a lot of content out there that may not be new, but it’s new
to the vast majority of people.
And I think that there is a huge opportunity with all of these
entities spinning off these cable networks for somebody to say, you
know what, for the right price, I would put these things together and
create an opportunity to, you know, I think we’re already starting to
see a move toward bundles, you know, where we used to pay for a bundle
from our, you know, Comcast, our DirecTV, our spectrum.
and we would pay for 500 channels and maybe we watched a dozen
channels. So, you know, people got fed up with that and they said,
hey, wouldn’t it be great if we got to a place where we only paid for
the things that we actually wanted to pay for? And I think that
streaming has brought that opportunity into reality. And where are we?
We’re actually saying, damn, I’m spending a lot of money on these
streamers, you know, between, you know, the five, six,
services that I have in order to get the most popular content, I’m
spending as much if not more than what I was spending. And now with
these spin-offs of these cable networks, think these corporations are
looking at them as a drag on earnings. mean, certainly many of them
still have tremendous revenue streams. They still have content and
franchises, things that can be spun off, things that can be…
monetized in other ways. To me, that’s something that I’m looking at.
And I’m saying, okay, you know, the person who figures out how to, you
know, for the right price, gather these assets up and offer them to
somebody, because we’re moving back to a bundle where, know, you’re
going to, know, Disney’s already doing it with Disney Plus and Hulu
and ESPN. You can buy them individually, but if you buy all three, you
can get them

monetized in other ways. To me, that’s something that I’m looking at.
And I’m saying, okay, you know, the person who figures out how to, you
know, for the right price, gather these assets up and offer them to
somebody, because we’re moving back to a bundle where, know, you’re
going to, know, Disney’s already doing it with Disney Plus and Hulu
and ESPN. You can buy them individually, but if you buy all three, you
can get them
for great price and now you’re starting to see Netflix with others,
know? And I think you’re, we’re gonna see more and more of that. But I
also feel like that there’s content out there that people have yet to
discover. And if they’re given a new opportunity to reach people that
they’re potentially gonna find an audience and bring more value to
whichever platform are those to them.
Orlando (49:12)
Is there, I mean, is there a silver bullet law that could be written
at the federal level that would both respect copyright and artists
while allowing for creative advancement of these tools?
Rob Rosenberg (49:25)
That’s a big question. I do think…
Orlando (49:27)
Right.
Or is it just gonna be
a mishmash of local decisions and state cases and things like that?
Rob Rosenberg (49:37)
Yeah, mean, earlier on, talked about where I mentioned compulsory
licensing and music, you know, so it’s a place where people can can
use music in certain ways, knowing that as long as they pay whatever
the statutory license is, you know, either to a collection agency like
an ASCAP, a BMI, which then distributes it to the creators, they
they know that they can move forward with confidence and certainty. I
think that there is probably a concern that if we move toward that
kind of model for training content, that certain types of content that
would really bring very high value in the marketplace, it would get
licensed for the same thing that an insignificant or less significant.
copyrighted work would bring. there may be, you know, some disgruntled
members of the content community unless there were, you know, you
know, A level content, B level content, et cetera, that either is
determined by, you know, how much money, you know, I’m thinking it’s
probably money, you know, how much that property has brought in
either, you know, through distribution, through licensing, et cetera,
before. But
But there might be a scheme like that that could be worked out if the
parties were willing that would give people certainty, would allow the
platforms to move forward, and would also make sure that there’s
compensation coming to those copyright holders.

Orlando (51:11)
⁓ Now, Rob, we have one question that we ask everybody here, and it’s
that, you know, ⁓ an old ⁓ boss of mine had this line where he said,
you’re useless to me unless you fucked up big at least twice. ⁓ And I
love that line because it implies that you need to have, you know,
you’re gonna be more valuable after the fuck up. Can you remember, and
would you be willing to admit, any time where you screwed up and what
you learned from it?
Rob Rosenberg (51:23)
Mm-hmm.
⁓
Yeah, I mean, on a high level, there was an instance where I had a

contract with an employer. So, you know, basically I had a non-
compete. I couldn’t go elsewhere. You know, I was hungry. I was eager

and ended up, but through circumstances, having some conversations
with the parent company of my employer. And, you know, they considered
hiring me away from my employer.
And they’re the parent company, so it wouldn’t be a violation of my
agreement. And I thought, all right, well, here’s a potential for me.
But I was very happy with the employer as it was. I was just looking
for more of whatever opportunities. when basically all of this came
out, there was a lot of unhappiness by my current company. ⁓
Orlando (52:26)
Bye.
Rob Rosenberg (52:27)
that I was disloyal, felt, you know, at the end of the day, what
actually happened is my current company said, you know what, we’re,
you know, we want to keep you. So we’re going to go and talk to the
parent company and say, you know, we don’t want you to take them. You
know, we want to keep them and we’ll make it worth as while to stay
with us. So it actually ended up being, you know, it ended up being a
net positive. But I think, you know,
that there were people who, know, they thought that, you know, I was
locked up tight because of an employment agreement. And, you know,
then having that conversation with the parent company, I think, you
know, turned a few heads and you had the potential to spoiling
something that otherwise, like I was really otherwise very happy with.
But, you know, it ended up strengthening all those relationships all
around. Look at that.
Orlando (53:13)
I’m surprised they didn’t see that as a good thing, know, somebody
advocating on their behalf.

Yeah, like I think that that would be a good thing. Like somebody who
we like and likes us is gonna go to the parent company, fabulous.
Rob Rosenberg (53:25)
Yes,
believe me, that’s how I pitched it. ⁓ you know, well, I think people,
know, everybody wants, if you got a good thing going, you don’t want
to share. you know, that’s fine. It was nice to know that they were
willing to go to bat for me.
Orlando (53:28)
Yeah
Yeah.
Well, I believe that they had a good thing going. You’re so
knowledgeable and so charming and it’s been so nice to chat to you.
It’s been great to meet you and I just, I think your perspective on
this is so valuable and thank you so much for talking to us. I really
appreciate it.
Rob Rosenberg (53:41)
Thank you.
Well, I’m happy to be here. Thanks very much for having me. And I do
have a sub stack that I call the Technotainment Scorecard. So if you
want to check it out, it’s a weekly column that I write on topics just
like this.
Orlando (54:09)
well, I’ll definitely join and I’ll make Audrey join as well. yeah,
we’ll also put it in when we post it, we’ll make sure that we link to
it so we can get a lot of people reading it on a weekly basis.
Rob Rosenberg (54:21)
Great. Thanks so much for talking to me.
Orlando (54:22)
Great, Rob. Well, thank you so much.
Absolutely. Take care.
Rob Rosenberg (54:26)
You too.