Studios Are Lousy at Predicting the Future
Why the WGA has the task of forecasting how AI will affect motion pictures
This is the third in a series of articles about the Writers Guild strike and the advent of artificial intelligence. Although the Writers Guild is relatively small, this union is grappling with issues that will soon affect the entire freelance workforce of 73 million.
Personal story: when I started my job as Vice President of Online Programming at Sony Pictures in the late 1990s, a legal box stuffed with thick contracts awaited me in my new office. I sifted through it, noticing that these agreements covered everything from technology vendors to individual employment deals. They were rich in specific detail. For instance, one producer’s contract stipulated the exact dimensions of his office: when I arrived to meet him, he had a measuring tape in hand, ready to show me that the studio was in breach of contract. Doing his best to make a good first impression!
The studio’s obsession with complicated contracts puzzled me. A colleague took me aside and explained, “There’s a reason why studio contracts are 100 pages long. It’s because they’ve invented so many ways to screw people.” Then she added, “No matter how carefully you draft your next contract, they will always find another way.”
Showdowns between the studios and the unions tend to be drawn-out affairs. The studios can withstand a short strike. They have plenty of notice before the contract runs out. That’s why they are well prepared.
This time around, the streaming video companies have stockpiled enough shows and films to make it through a dry spell. Broadcasters and movie studios will be able to fill the airwaves, theaters, and internet for a while. A few programs, like the late-night talk shows with comic hosts, have gone dark and must now rely on reruns. But the most video programming services won’t look much different next week or next month.
After a few months, however, as the sound stages remain vacant and the video distribution pipeline gradually runs dry, the pressure on the media companies will begin to increase. That’s when they will return to the bargaining table with a meaningful offer.
To prevail, the writers must be mentally prepared for a siege. A WGA strike usually means a grueling standoff of 100 days or more. The writers will devour their savings while waiting for the studios to concede. It is a test of wills and endurance.
Moreover, the other unions who support the WGA by refusing to cross a picket line will also suffer a loss of income. The longer the strike goes on, the more stress will bear down on solidarity. There may be a breaking point. If the solidarity between unions fractures, then the writers will lose significant leverage.
At the center of this dispute is a contract known as the Minimum Basic Agreement. As the name suggests, the MBA sets forth the absolute rock bottom deal points for any studio that wishes to engage a professional screenwriter.
The MBA is the floor, intended to prevent the studios from undercutting the basic terms. There is no ceiling: the job of the talent agents is to test and probe continuously at the upper boundary to find ways to extract a better deal. As I wrote yesterday, this arrangement works because it provides a minimum baseline income for most writers while preserving the opportunity for the lucky or especially talented few to earn significantly more.
If all other factors held steady, a well-crafted MBA could keep the peace for a long time. But other factors never hold steady. That’s not how the future unfolds. Rapidly-evolving digital technology introduces a wildcard that can tip the balance in favor of the studios.
Whenever a new distribution platform emerges, for instance, the studios tend to reap windfall profits while sharing only the bare minimum with the writers and other talent.
The studios manage to get away with this because the MBA is typically mute (or deliberately vague) on new technology. That’s why the burden falls heavily on the Writers Guild to set forth a clear vision of how new technology will unfold and precisely what the studios can and cannot do with it.
In sum, the task of being an exceptionally good forecaster falls on the WGA.
That’s tough for the writers. But it makes sense because movie studios are, candidly, terrible at anticipating future trends.
For 25 years in my career, I have dealt with big media companies in a variety of roles: as an executive in a movie studio and in TV networks; as a distribution partner for video entertainment; as a licensee who activated media brands as games; as a partner in new startup ventures; and as a strategist advising certain media companies on long range scenario planning.
These experiences taught me three things:
media executives tend to reflexively resist new technology,
they rarely bother to learn how new technology works, and
they are horrendously bad at anticipating how it will impact their business.
Too extreme? An abundance of examples support this conclusion. The most famous flub was MPAA President Jack Valenti’s dire forecast that the VCR would be like Boston Strangler to the movie business, causing it to “bleed and bleed and hemmorage.” Valenti wasn’t just wrong. He was incredibly wrong. Home video soon emerged as the most reliable and most profitable revenue stream. This bounty lasted for decades after Valenti’s forecast.
Similarly dire predictions were made about the advent DVDs and the DVR and on-demand video: all wrong. Each successive technology was viewed as an existential threat to the movie business, but that never happened. Instead, each invention expanded the market for video entertainment.
Even YouTube was greeted with lawsuits until the studios and TV networks finally figured out how to use social media to drive views and awareness and eventually even to monetize their short clips.
That’s a dismal track record at forecasting, but there’s an even bigger blind spot. The media companies routinely fail to recognize new opportunities and that’s why they always fail to capitalize on new technology at an early stage when they might seize the leadership position.
Again, this claim can be backed up by examples from every era of film and TV. The mainline movie studios resisted the advent of television in the early days after World War Two, leaving the market open to newcomers like Columbia. Then the big broadcast networks eschewed cable TV in the 1970s, leaving the market wide open to newcomers like HBO (part of the Time Magazine empire), Turner (local TV stations) and Viacom (a drive-in movie theater chain!) to dominate.
The cable TV giants grew no wiser with age: Viacom invested in moribund Blockbuster Video around the same time that HBO passed on the chance to acquire Netflix. The entire Big Media Industrial Complex failed completely to understand streaming media, equating it with YouTube and low-quality. Instead of embracing the open Internet, the media giants doubled down on the walled garden of cable TV with a half-baked scheme called TV Everywhere. That project was a spectacular flop. It cost the networks and cable giants years of distraction, which left the streaming market wide open for Netflix and Amazon, who proceeded to define it.
More recently, the broadcasters and movie studios were completely oblivious to live streaming until Amazon swooped in and acquired Twitch for $1 billion. Ditto TikTok.
This habitual failure to spot the Next Big Thing for decade after decade is quite costly. Hollywood’s strength, power and prestige has ebbed. Big media now operates at a scale disadvantage relative to the technology giants of Silicon Valley and Seattle.
Which explains why Rupert Murdoch pulled the ripcord and bailed out of the filmed entertainment game by offloading 21st Century Fox to Disney in 2017.
After decades of missed opportunity, the most famous movie studios are weaker:
Walt Disney has been supersized, gobbling up the biggest and best brands, including Star Wars and Marvel Entertainment in addition to Fox. Disney is bleeding cash to grow its streaming business in its attempt to catch up with Netflix.
NBC Universal is owned by the cable colossus Comcast, which also rolled up Dreamworks.
Columbia and Tristar were flipped by Coca Cola to a Japanese electronics company, where they were rechristened Sony Pictures Entertainment.
MGM was swallowed up by online retailer Amazon.
Paramount and CBS were first spun out, then reacquired, by Viacom, in a bizarre sequence that made sense only to magnate Sumner Redstone. Now the remnants have been renamed Paramount Global.
TimeWarner, including the legendary Warner Bros. studio, was absorbed by AT&T in the most misguided acquisition since, well, … that time when AOL acquired TimeWarner. After a couple of years and a new CEO, AT&T woke up with a hangover, regurgitated the half-digested WarnerMedia and forced the zombielike shell into a shotgun wedding with Discovery Communications, conferring $70 billion of debt as a dowry. The combined Warner Bros. Discovery is now preoccupied with making debt payments at a time while rivals are investing massively in streaming media.
Money was made, the top executives cashed out, but only one of these companies is stronger today than it was a decade ago, and even that is debatable.
Look how tiny the movie studios are compared to Apple, Microsoft, Alphabet and Amazon. And Meta, ByteDance, Tencent, Alibaba and Reliance.
There are many reasons for this failure to anticipate technology trends. First, speculating on technology is not in their wheelhouse. Movie studios and TV networks are not venture capital firms. They prefer to apply their speculative capital to new shows and movie franchises, which is closer to their comfort zone, and it is something that the big studios have done successfully for 100 years.
Second, these companies are disconnected from their end consumers: putting content in front of audiences is a job for movie theaters, cable operators, distributors, and retailers. As a result, the big media companies do not really have a finger on the pulse of consumer sentiment. Which is a paradox considering how much influence they wield over the public imagination.
But perhaps the main reason these firms fail to anticipate the future is that media executives just don’t care very much about technology.
The most innovative technologies that have been successfully deployed in the motion picture industry came not from executive fiat but at the initiative of creative artists, such as legendary animators like Walt Disney, Harry Harryhausen, and Ed Catmull, directors George Lucas and James Cameron and even puppeteers like Gerry Anderson and Jim Henson. These artists invented entire categories of special effects to realize their creative visions.
Technological innovations in distribution came from cable maverick Ted Turner and satellite pioneer Charlie Ergen. Outsiders, never insiders.
This history of resisting or ignoring technology trends puts the studios at a disadvantage today, because digital media is in the process of reconfiguring every aspect of the filmmaking process:
Generative AI promises to compress and speed up the first stage of ideation and exploration and storyboarding and concept development.
Virtual production using game engines and LED walls has already begun to collapse the long cycle of video post-production and special effects, compressing it into the principal photography phase of production. Filmmakers tell me that they can achieve 50% cost reduction or reduce shoot days by 2/3 using virtual production.
Video distribution outlets have proliferated since the pandemic. Content from individual influencers dominates these new video outlets. Even Twitter is distributing TV shows now. The traditional media companies do not dominate these new platforms as they did broadcast television, cable and satellite.
The motion picture companies are not the authors of this transformation; they are subject to it. They are on the receiving end of technological change. No doubt, they will eventually find a way to eke out some advantage from these innovations, but they are not in control of the process.
Given their dismal track record of forecasting the future trajectory of technology, it comes as no surprise that the media companies prefer to hedge their bets when it comes to writing a contract that involves technology.
After all, what is a contract but a forecast about future behavior?
The preceding commentary is an embarrassingly long windup to what I confess is a modest point: when it comes to negotiating a new MBA with the Writers Guild, the movie studios prefer to give themselves maximum wiggle room on the topic of AI.
During previous negotiations, when it came to emerging digital technology (then known as “New Media”) the studio negotiators routinely pleaded “it’s much too early to predict what will happen” and refused to make clear commitments. At the time, the writers were focused on locking in the terms they had negotiated for traditional broadcasting and film. Rather than risk those gains by overnegotiating, the WGA accepted the studio’s fuzzy position on new media.
The failure to pin down the specifics of new media in contracts backfired on the writers. The studios soon raked in windfall profits on home video, particularly during the rapid rise of DVD in the 2000s.
Then in the next round of dealmaking in 2007, while the WGA was preoccupied trying to fix the home video mistake, the studios did it again by carving out wiggle room around then-nascent streaming video: in the ensuing decade, streaming video soared and began to displace traditional forms of distribution and crush home video. Once again, the writers got bilked.
This time around, the writers have sharpened their axes. They won’t be fooled again. They are determined to anticipate new technologies and plug up any possible technological loopholes that might emerge in the future.
Meanwhile the producers plan to stick their usual approach during negotiation, namely, “Who knows what might happen? Let’s cross that bridge when we come to it.”
This puts the burden of forecasting the future of technology on the WGA.
The specific demands of the screenwriters can be grouped into a few categories, each of which has a particular technology focus. The WGA released a high level summary of their demands in the current contract dispute, which include:
1. Fair payment across all platforms:
Increase minimum compensation to address the devaluation of writing in all areas of television, new media and features;
Standardize compensation and residual terms for features whether released theatrically or on streaming;
Ensure appropriate television series writing compensation throughout entire process of pre-production, production and post-production;
Increase residuals for under-compensated reuse markets;
Restrict uncompensated use of excerpts;
Apply MBA minimums to comedy-variety programs made for new media
2. Better benefits and working conditions and fair treatment:
Increase contributions to Pension Plan and Health Fund;
Address the abuses of mini-rooms;
Enact measures to combat discrimination and harassment and to promote pay equity
The writers argue that their salaries have not gone up in a decade. With inflation they are working for about ~15% less than they did ten years ago. On top of that, they are often booked on shorter projects now, which means they work less in total each year.
Film and TV producers are also hiring smaller teams of writers, called "mini rooms", which means that writers get less work and less pay despite the decade-long boom in streaming video. The term “mini room” sounds innocuous but what it really means is a reduced number of writers generating lots of new ideas in a much shorter amount of time. That’s a lousy deal. This outcome has been compared to a “gig economy” for screenwriters: short term employment at minimal rates.
The studios abused the fuzzy language in the contract to weasel out of their minimum basic commitments in the MBA.
Notice in the points above that the WGA is focused on addressing the imbalance caused by technology-driven innovations that they missed in the previous contract: new distribution systems (streaming) and new formats for new media. Taken altogether, these provide a vivid example of how a disruptive cycle of technology-driven change (such as the advent of on-demand video streaming over the Internet) leads to the redistribution of profit. In this case, the money has been redistributed away from the writers: overall budgets are up but the amount spent on writing is down.
But all of that is about cleansing the sins of the past.
The deal point that most concerns this newsletter is the third, which is all about the future.
3. Guarantees about the use of artificial intelligence
The WGA does not seek an outright ban on artificial intelligence but rather an agreement to “regulate use of artificial intelligence on MBA-covered projects: AI can’t write or rewrite literary material; can’t be used as source material; and MBA-covered material can’t be used to train AI”.
In some press reports, this demand has been framed as a total ban on AI, but that claim is entirely bogus. The WGA is not seeking to deny the use of AI completely.
Instead, their demand is carefully calibrated to prevent the use of AI to degrade the rights of human writers and reduce their fees. As the WGA stated:
“The WGA’s proposal to regulate use of material produced using artificial intelligence or similar technologies ensures the Companies can’t use AI to undermine writers’ working standards, including compensation, residuals, separated rights and credits.”
On the opposite side of the bargaining table sit the representatives from the Alliance of Motion Picture and Television Producers. Founded in 1982, this trade association represents the interests of approximately 350 member companies during collective bargaining sessions with the entertainment guilds and unions.
Members of AMPTP include the major motion picture studios (Walt Disney Studios, Warner Bros., Paramount Pictures, Sony Pictures, MGM, Universal Pictures), the broadcast television networks (such as ABC, CBS, NBC and Fox), and the streaming video platforms (Netflix, Apple TV+, Amazon Prime) as well as several cable TV channels and independent production companies.
Given their lousy track record at forecasting technology trends, it makes sense that the movie studios would prefer to avoid getting pinned down about the use of AI.
Besides, in previous negotiations they reaped a massive windfall by leaving future technology commitments fuzzy and undefined.
So when the negotiation turned to the topic of artificial intelligence, they once again opted for wiggle room.
Here’s how they did that: gracelessly.
The AMPTP flatly rejected the writer’s demands concerning AI. They did not even engage with the substance of the writers’ complaints. Instead, the best the studio heads could muster is a purposefully vague offer to attend “annual meetings to discuss advancements in technology.”
Seriously, that’s it.
Annual meetings to discuss AI.
No commitment whatsoever.
No wonder this lame proposal fell flat on its face. And immediately thereafter, the strike began.
One writer described it to me like a scene from a movie: “Imagine our negotiation with the AMPTP as if there were two people who are planning to get married. While the couple is drafting the terms of a pre-nuptial agreement, one spouse-to-be proposes a clause that stipulates that both parties will remain faithful and honest to each other. And the other spouse replies, “It’s way too early to commit to that, darling. Instead, let’s have a meeting once a year to see how it’s going.”
The AMPTP is telegraphing the studios’ intent. Their non-response makes it obvious that they intend to introduce generative AI into screenwriting. By some accounts they are already using it.
The studio honchos know that AI is improving fast. If they can hold the writers at bay for just one year, then that will give them plenty of time to establish something like an improved version of ChatGPT in the creative process.
Once AI is in, it will never be removed. ChatGPT is the thin end of the wedge.
The writers know this. That why they are dead set against letting the producers slide AI into the workflow without a fight. If it happens at all, the writers insist, it will be done by the writers themselves on their own terms.
Given the intense economic pressure on the studios and TV networks from an evolving competitive field in a completely redefined distribution landscape, it is a certainty that the producers will seek to use every means at their disposal to obtain more writing for less money. Which explains why they are so keen to use ChatGPT and similar systems to generate screenplays, even if the output is inferior to a human writer.
Easy prediction: someday, some professional writers will be boosted by GPT4 or GPT5 or another future version serving as a co-pilot or automated writing assistant. These writers will gain superpowers, for instance the ability to deliver more scripts faster with fewer human writers in the room. The net result will be: less employment for non-augmented writers, and plenty of work for just a few superhuman writers who are equipped with AI. This scenario is an affront to the equity principles of the WGA.
What’s really at stake in this dispute? Is this just a struggle for control? Artistic principles? Containing cost? Maximizing profits?
No doubt, each of these is a factor in the dispute, but at the core, this is the struggle to determine who will be the creative progenitor of motion picture projects.
In film, the “creator” credit has long been the subject of a debate, a kind of tug-of-war between directors and writers.
But in television (and streaming video) it is a settled matter: the writer is the creator of the series. Television is a writer’s medium.
But wait! Frequently, the producer is the person who comes up with the initial concept and often it is the producer who provides the impetus to get the film or TV series made. Could AI give the producer the opportunity to claim authorship, or at least dislodge the writer? Possibly.
Which is exactly why the gnarliest bone of contention is about rewriting content generated by AI. Specifically, the WGA refuses to accept any script written by a machine as “literary material” and according to the WGA proposal, screenwriters cannot be hired to rewrite or polish a rough draft generated by an AI like ChatGPT.
What this tells us that the WGA has better much futurists in its ranks than the movie studios.
The WGA has carefully considered the worst-case scenario. It goes like this: in the future, a movie producer will talk to GPT-12 on the drive into town from Malibu. He will ask the AI to generate a script that features two famous stars in a buddy cop movie intended for summer release, written in the style of “Bad Boys”. Or a dystopian thriller set in an aging space station with a dangerously unreliable nuclear core, written in the style of “Alien”. Or some such formulaic tripe. All barked into a mobile phone by a short man driving a big car too fast on the PCH. And then when he arrives at his bungalow on the lot, he will have in hand a complete script auto-generated by the AI. And then, to complete this worst-case scenario, that’s when the producer will call in a WGA writer, slide the AI-generated script across his desk, and tell her, “I’ve got this screenplay, it’s a star vehicle for summer release, and I want to hire you to do a re-write.”
That’s the worst case scenario for screenwriters, not just because they will be paid a fraction of union scale for a rewrite, but more importantly because it will signal the death of human authorship.
I know it may seem like I am exaggerating this scenario, but this is precisely what concerns the writers. As one writer told Rolling Stone:
“I think with the exponential growth we’re seeing, it’s not outside the realm of possibility that this thing can spit out a screenplay in a couple of years, and the question will become: ‘Do you need to hire a writer at their quote to do that first draft, or do you give them the rough draft from ChatGPT and tell them to clean it up?’ Hopefully, that doesn’t become a reality, but better to get a handle on it now than to regret it years from now.”
For the studios, this is mostly about carving out some wiggle room so that they can chisel a few more nickels from the artists who create the grand narratives that propel the entire industry.
For writers, this is an existential threat.
So this explains the standoff in the negotiations over AI. That said, my impression is that both parties are thinking about AI the wrong way, for very different reasons. In my next newsletter, I’ll tell you why I think that generative artificial intelligence is likely to play out in very unexpected ways for the media business and other industries. If you’d haven’t read the previous articles in this series, why not check them out in the archive?
Ok so if the WGA strike a deal to limit the role of AI in screen writing, will other producer countries without such limitations gain a competitive advantage?