AI is already beginning to be deployed in some areas of the film and TV production process,1 though the potential magnitude of its long-term impact is still coming into focus. Our research and discussions with studio executives, producers, and technology leaders suggest that uncertainty around AI extends beyond whether and how it will change production to how those changes manifest throughout the content and distribution ecosystem. While the technology’s limits, adoption trajectory, and potential scale of impact are yet to be determined, historical technological shifts and early use cases suggest AI could, over time, materially alter the industry’s structure and profit pools.
As a result, industry leaders face practical questions about near-term operating choices and strategic questions about what AI could mean for their businesses longer term.2 Based on our recent experience, research, and discussions, AI’s expanding capabilities have prompted some leaders to begin to reevaluate their business strategies while recognizing they must also manage looming concerns about labor impacts, potential risks, and the nature of creativity. As Sean Bailey, an industry veteran and founder of B5 Studios described the challenge to us, AI may represent “a more significant platform shift than we have ever seen before in our industry.”
To understand how AI could impact the overall video content industry, we interviewed over 20 media leaders, including studio and production executives, talent agents, AI innovators, and academics; drew learnings from our work with video content companies; and analyzed broader industry data and the history of technology innovations in content production and distribution (see sidebar, “About the research”).
The rise in AI comes at a moment when video industry players are already under immense pressure. Consumer attention is fragmented amid an abundance of content and finite viewing time, and attention is shifting away from traditional channels to streaming platforms and user-generated content (UGC). In the United States, for instance, daily viewing hours spent on linear TV3 declined by 4 percent CAGR from 2022 to 2024, while streaming4 grew by 13 percent and social video platforms grew by 14 percent (Exhibit 1). Consumers are also changing how they watch, consuming video on mobile devices and increasingly using their TVs to search for and view online videos, including user-generated content.5
At the same time, investment in content is leveling off. In the United States, which represents over half of global spend,6 original content spend is projected to decline by 2 percent per year as buyers turn to sports rights and licensed programming, which can attract outsized audiences or cost less (Exhibit 2). These projections do not include the potential impact of AI, which introduces significant uncertainty around content spending.
Amid these shifts in supply and demand, leaders interviewed indicated AI has the potential to impact many production processes,7 as well as back-office operations.8 New tools and early experimentation are already demonstrating single-digit productivity improvement potential in some use cases, while also raising important questions about intellectual property (IP), authenticity, and the future of creative work.9
While the longer-term implications are not yet clear, this research seeks to identify some potential industry outcomes, starting with already-emerging impacts on production workflows. It also explores how AI could change what content is created and who creates and distributes it over the longer term under varying adoption scenarios.
The possible outcomes discussed include the democratization of high-end content creation, accelerating the shift of consumer attention from professionally produced content to UGC platforms and smaller studios, as well as entirely new forms of content and distribution, including more immersive, personalized, or participatory entertainment. Our report also identifies the potential redistribution of economic value and the potential for a net increase in total content supply and demand if AI’s impact is similar to past disruptions from the rise of new production and distribution technologies. This includes potential impacts on production workflows, influencing approximately 20 percent of original content spend in the next five years10 and redistributing up to $60 billion of annual revenue within five years of mass adoption of AI use surpassing that of incumbent technology.11
Many in the industry have raised concerns about how AI adoption could affect these potential outcomes and other possible scenarios, including the impact of AI on creativity and the already-declining number of entertainment jobs, while others question whether certain uses of AI should be permitted at all in creative production. Current adoption and impact is uneven, and its future trajectory is uncertain. As several executives made clear, AI-generated output is not yet at a quality level to drive meaningful disruption, with content that in many cases does not meet premium production standards. While this may change, leaders interviewed noted that there are limits to how deeply AI will disrupt end-to-end content creation and the art of storytelling in established TV and film formats.
This report is limited to exploring the business, not creative, implications of AI, and it does not assess the choices that could arise as the impact becomes clearer (for example, how best to take advantage of the productivity improvements AI may enable or the demand for new creative roles and skills even if existing ones are affected).
In addition, several studios have challenged relationships with foundational models and publicly available text-to-video tools that they claim were illegally trained on the studios’ IP. And there are significant unresolved issues and debates around authorship and rights ownership, especially from members of the creative community.
With all these forces at work, many critical uncertainties remain about the impact of AI on content production, innovation, and the longer-term distribution of economic value across the ecosystem. In the absence of clear answers, the insights that follow are intended to provide industry leaders with a foundation for assessing what it may mean for their businesses operationally and strategically, what markers to monitor, and what questions to ask as AI capabilities continue to develop.
AI is already showing potential to reshape some core production activities
Leaders we interviewed report experimenting across select production processes and finding potential for 5 to 10 percent productivity increases in specific use cases this year.12 Much of this initial value comes from trials across development and pre-production workflows, with leaders telling us they expect to expand to physical production and post-production over the next five years.
Development and pre-production at the forefront
Development and pre-production can be particularly time-consuming, especially when communicating creative vision. Leaders noted that, while it is still early, more developed use cases exist in this area than in other production phases, in part because they require less technical complexity and pose lower adoption risk, making it a logical starting point. Indeed, our research finds that these early adopters are already realizing mid-single-digit productivity-related increases across the pre-production process, with more in certain genres or workflows. “People are now coming to pitches with clearer visual directions and pre-production steps [using AI],” said a former executive at a streaming company. Leaders interviewed noted longer-term potential for pre-production changes to affect later stages, such as physical production and post-production. This could include a mindset shift from the norm of “fix it in post” to “fix it in pre,” according to Hannah Elsakr, Adobe’s vice president of gen AI new ventures. Similarly, use cases including A/B testing of story beats and analytics-led script breakdowns that identify production elements, such as location, props, and shot lists, could allow teams to be more targeted and shoot faster when on set.
Room for advancement in physical production
Physical production, including set construction, principal photography, and reshoots, represents the majority of production activity.13 Despite some examples of AI use in physical production, such as Netflix’s first original series with AI-generated final footage,14 early adoption lags that of pre- and post-production. Adoption barriers cited by industry executives include limitations of current tools; protections for creative talent across IP, contracts, and labor; and concerns around consumer preferences. “Creative integrity is so important,” said Kevin Lingley, executive vice president of AI at Fremantle. “It doesn’t matter what is used to produce content; viewers need to feel they are being engaged and entertained in the best way possible, and not being lied to.”
If these challenges are addressed, interviewees identified potential for two long-term shifts. First, on-location and shooting processes could change. “Think about car chases, where you historically had to lock down [parts of] a city for two months,” said a former executive of major film studio. “Instead, we could re-create locations virtually.”
Second, physical production could be accelerated, both by shifting work into pre-production and by reducing reshoots. “Shorter production cycles are a huge advantage to maintain momentum in a hit-driven business,” explained Jan Lacher, senior vice president of content and business development, RTL Group.
Usage growing in post-production
Post-production work is often outsourced and can be the second most intensive step, particularly for VFX-heavy sci-fi and blockbuster movies.15 Producers noted that AI is already used to dub and localize content and to accelerate footage clipping and video library filtering, a trend illustrated by Moments Lab’s partnership with Banijay.16 Many highlighted that additional uses are still to come. For example, a former executive at a major media company projected that once AI is adopted, it could accelerate current processes and eventually define new AI-native workflows in animation and VFX projects. Once tools reach what RTL’s Jan Lacher called “professional-grade resolution and consistency,” post schedules could shorten significantly, according to leaders, and AI could start to blend post-production into the pre-production process. “That is revolutionary in terms of the way we make content,” said Adrienne Lahens, CEO and cofounder of Infinite Studios.
What past innovations may indicate about AI’s potential industry impact
While there are early signs of impact as organizations experiment with opportunities in content creation, the longer-term implications for the overall industry remain even more uncertain, with multiple possible outcomes.
To understand how the industry might evolve, depending on the extent of AI adoption, our research explored what ten of the most significant past content production and distribution technology innovations17 might teach us about AI’s potential (Exhibit 3). Four main lessons emerge from this analysis:
- New production technology drove significant change in industry economics, redirecting some spend to tech vendors and passing most of the value to distributors. The introduction of digital cinematography illustrates this pattern. It marked a shift from film stock and processing costs to digital production and enabled faster, more flexible workflows on-set.18 Some of the value delivered was captured by the vendors who provided this new technology, but in many cases distributors gained the majority of the value at stake from production technology shifts through higher profit margins.19 A similar redistribution of value could occur with AI as a result of the structural fragmentation of producers versus concentration of distributors, with seven buyers making up 84 percent of US content spend.20
- Historically, most value from production innovations accrued to large incumbents with the means to invest, while consumers have benefited in multiple ways. For example, the first cameras were accessible only to large professional studios,21 and computer-generated imagery (CGI) expanded highly expensive effects-driven genres like sci-fi and fantasy.22 CGI improved the visual fidelity of films, raising the bar for premium productions, and average Hollywood blockbuster budgets have increased by 30 percent over the past two decades, with CGI shaping the highest-budget films.23 Home video cameras, PCs, and mobile phones have put video capture and editing tools in the hands of consumers, but AI may be the first technology with the potential to enable creation of content at higher production values outside of the large professional studios if tools improve sufficiently.
- Most distribution innovations expanded consumer choice and often increased overall content supply and demand. Television opened up more viewing occasions and time to watch and required more video content to fill the available capacity.24 The VCR and DVD made it possible to both capture and view films on demand at home, creating more control and more choice, including the rise of “direct-to-video” films.25 UGC platforms enabled access to an abundance of free-to-watch, on-demand content across niche genres and formats.26 Consumers now watch hours of user-generated content per day (see Exhibit 1).
- New technologies are often adopted in unexpected ways, giving rise to new content formats and distribution models. For example, early filmmakers treated the camera as a tool for recording stage performances, rather than the basis for new forms of storytelling that it became.27 Television programming first focused on transmission of live events and education lectures yet went on to support new episodic content formats.28 Leaders noted that the use of computers in the animation genre was first considered “niche” until Pixar evolved into one of the largest animation studios. Most recently, mobile-phone cameras played a significant, if not primary, role in the rise of short-form content formats and the open distribution platforms that are major destinations for consumers and advertisers.
Would you like to learn more about QuantumBlack, AI by McKinsey?
Three potential ways industry dynamics could evolve
Together, these lessons from past innovations and our broader research suggest three potential industry-wide outcomes from the scaling of AI: scaling of changes to current production workflows, wide-scale democratization of professional-grade content creation, and the creation of new content formats and distribution channels. These outcomes, which are not intended to be exhaustive or mutually exclusive, are presented in order of their likelihood of occurring, based on our expert interviews and analysis. Whether and how they ultimately play out (or other outcomes take shape) will depend on the extent to which high-end AI production tools become democratized (and who captures the resulting value), creatives adopt these tools across workflow steps, consumer viewing preferences change, new formats develop, and new distributors and aggregators emerge.
1. Scaling of changes to current production workflows
Leaders we interviewed noted that it is likely AI will increasingly impact production workflows, given the early signs their organizations have already seen. “I looked at every step of the workflow from ideation to distribution, and I really think every single piece of it will be significantly disrupted,” said B5 Studios’ Sean Bailey. Our analysis indicates that approximately $10 billion of forecast US original content spend in 2030 could be addressable by some form of AI.29
While adoption remains limited today, leaders interviewed noted a lag between technological development and industry-wide uptake, as seen in the historical analysis. CGI provides a clear precedent. As highlighted by Bryn Mooser, cofounder of Asteria, Jurassic Park’s landmark effects “changed the entire industry,” ultimately increasing film budgets, even though Lucasfilm’s Industrial Light & Magic had been founded nearly two decades earlier. However, interviewees said industry players will need to address the many concerns about the technology’s use in artistic and creative endeavors, such as how it affects creativity or authorship.
Interviews and past historical lessons suggest distributors are positioned to capture the majority of value delivered from AI-driven increases in workflow speed and capacity. This is driven by structural market dynamics, including a crowded producer market, consolidating buyer landscape, and transparency of production budgets.30 Producers who make investments in new technologies, adapt their operating models, and have or develop strong IP are also well positioned to capture a portion of this value.
How much economic value technology vendors could capture remains unclear. If broader video tools follow the trajectory of large language models, with intense competition between vendors and open-source models closely trailing best-in-class offerings, then value may largely pass through to distributors and producers as tools become democratized and inexpensive. However, if top models meaningfully diverge from others, tech vendors could adjust their pricing, capturing a greater share of value. Leading model developers and top IP owners could also form exclusive partnerships for professional-grade video models, building a closed-loop ecosystem, while traditional production service providers, such as motion capture providers or special effects (SFX) specialists, may likely face mounting pressure as routine tasks become automated.
2. Wide-scale democratization of professional-grade content creation
AI could also enable smaller studios and creative entrepreneurs to compete more directly with large studios, with the potential to increase total content supply and open up new opportunities for the broader creative community. The research indicates this outcome is less certain than the anticipated impact across existing production processes. It may depend in part on whether the smaller creators apply AI to deliver more high-end content rather than simply to increase the volume of content they produce. Already, early use of AI tools has led to the development of what many refer to as AI “slop.” Interviewees also acknowledged that incumbents still have distribution advantages. “In a world with even more content, IP owners will have a relatively higher chance of success,” said Michael Porter, an industry expert. “Their brands are more likely to cut through the noise in a world saturated with content.”
If widespread democratization of professional-grade content does occur, it could put additional financial pressure on traditional producers and distributors. Large incumbents would need to rewire long-standing workflows and operating models as this happens, while newer studios could adapt more quickly. Entrepreneurial AI-enabled creators and artists could tell their stories in new ways and open distribution platforms. “There are an ever-increasing number of entry points for content,” said Matthew Wilson, chief legal officer, Fremantle. “AI only adds to our ability to play across new distribution channels.” The combination of finite consumer attention and more content across more distribution points could have a significant impact on the current distribution landscape. A potentially analogous disruption followed the rise of broadcast TV, which contributed to a 38 percent drop in the number of cinemas between 1930 and 1957 in the United States.31
Even a modest shift in viewer behavior could have a meaningful impact on industry profit pools. If, for example, existing open video distribution platforms captured an incremental 5 percent of TV and film viewing hours in the United States, there would be a $13.2 billion decline in US TV and film distribution revenues. This loss would only be partially offset by a $7.5 billion increase in open-platform revenues.32 The roughly $5.7 billion decrease in net revenue in this example reflects a shift in viewership to open platforms with lower monetization per hour.33
3. The creation of new content formats and distribution channels
To the extent that AI gives rise to new content formats and distribution platforms, not just shifting distribution across existing platforms, it could significantly redistribute economic value pools among industry participants. The shifts from stage plays to cinema, linear to streaming, and long-form to short-form each contracted incumbent revenue by an average of 35 percent in the five years after the technology was widely adopted,34 with consumers receiving a wider selection of less expensive content.35 While AI is still early in its adoption curve, applying this historical pattern to forecast revenue indicates that around $60 billion of revenue could be redistributed within five years of reaching mass adoption, if AI use surpasses that of incumbent technology.36
Specifically, what these new formats could look like is difficult to assess, given the nature of the creative process. However, many executives said such a shift is possible, based on past experiences. “We will see a wave of new formats with AI, in the same way that nonlinear editing innovations gave rise to a wave of talent and reality shows,” said RTL’s Jan Lacher.
Dani Van de Sande, founder and CEO of Artist and the Machine, noted more transformative shifts could come as AI moves toward world models, systems that don’t just generate isolated assets but maintain an internal understanding of characters, environments, rules, and cause and effect over time. “As world models mature, we’ll see entirely new creative operating models and mediums, stories that persist across formats, characters that evolve beyond a single script, and narrative experiences that can respond to audiences or unfold differently over time ... . It’s redefining what a story is and how it’s experienced,” she said.
From here, numerous structural shifts could arise if historical patterns hold.
Fundamentally new models could emerge, including integrated platforms that combine creation and distribution in one environment, similar to how TikTok and CapCut enable AI-supported creation and near-zero-cost global reach.37 “Creators will move upstream as AI puts cinematic quality production tools in the hands of people who never had access to traditional Hollywood pipelines,” said Infinite Studios’ Adrienne Lahens. George Strompolos, co-founder and CEO of Promise, echoed that sentiment: “You could say that the creator economy was about the democratization of distribution. This is about the democratization of production and creation itself.”
One early example is DreamFlare, a hybrid creation and distribution platform where creators can publish episodic, AI-enhanced visual stories directly to audiences, who can vote on which concepts are developed into full shows. This could enable new storytelling formats beyond today’s conventions.38
AI could further reshape audience behavior and experience. Streaming platforms have already moved audiences away from a monoculture, in part by learning user preferences and dramatically expanding the supply of content and range of individual choice and control, a transition that further AI-driven hyper-personalization could accelerate.39 “One [shift] is that the user is driving the content that they want to see,” said Michelle Kwon, head of operations and partnerships at Runway. “And that means [in the future] that you’re no longer beholden to a single platform or season two of your favorite show not coming out for another year and a half. You’re able to generate a story with your own favorite characters or the ideas that you have in your head immediately today.”
Even as technology reshapes the production process, leaders emphasize the value of creative talent and taste, citing premium content anchored in human-led storytelling as a defining marker of quality. “What is most exciting [about AI] is the potential to expand our human imagination and get to a point in storytelling where we don’t have to be thinking: Can I shoot that on a film set?” said Lori McCreary, producer and CEO of Revelations Entertainment and a board member of the Producers Guild of America.
Those building video, image, and text models emphasized that human-led storytelling could also influence how the technology develops. For example, Runway’s Michelle Kwon envisions an integration among tool vendors into unified creative platforms. “There won’t be LLM companies and video generation companies,” she said. “The whole industry is moving toward world models.”
Ethical and risk considerations raised in our interviews
However the industry evolves, AI’s expanding role in content creation brings risks, many of which are already here. Across our interviews, three concerns emerged most consistently: talent and creative implications, IP and other rights infringement, and potential hallucinations or bias in model outputs. Leaders interviewed pointed to the critical role of regulatory and ethical frameworks, including measures such as training models on IP-safe content, standardizing AI use compensation, and ensuring creator consent with meaningful control over digital likeness.
Talent and creative implications
Creative erosion and related talent implications center on whether AI-generated or AI-modified work can reflect artistic intent and accurately portray lived experiences without distortion, as well as on AI’s ultimate impact on entertainment jobs. Guilds and unions have negotiated for language that protects artists.40 Recent events highlighting the sensitivity of these issues include the SAG AFTRA (Screen Actors Guild–American Federation of Television and Radio Artists) and WGA (Writers Guild of America) strike41 and debates over the movie The Brutalist, which used AI dubbing and voice modification to create more realistic Hungarian accents.42 The rise of digital likenesses adds complexity for talent, producers, and distributors trying to ensure fair compensation and protect against deepfakes.
Talent agencies, another key stakeholder, may create new service lines helping clients manage, monetize, and protect digital likeness, voice, and IP rights at scale, even if changes to production workflows disrupt their traditional services model. “We are focused on setting perimeters and protections for talent, IP owners, and rights holders,” said Alexandra Shannon, an executive at Creative Artists Agency (CAA). “Premium, authentic, human-led content will be even more valuable if the overall supply of content increases.”
IP and other rights infringement
Concerns regarding IP and other rights infringement have led to lawsuits over training models that have used existing intellectual property.43 These issues are prompting industry leaders to consider what Hannah Elaskr referred to as the “nutrition label” for model-training data. They are also giving rise to new partnerships between studios and AI companies to develop IP-protected proprietary models that are trained only on licensed data.44 “We have new types of legalities to worry about as producers,” said Revelations Entertainment’s Lori McCreary. “Now I have to ask if AI was used to create it [a script] to document where the human creative input came in and show that we can copyright the material.”
The risk of hallucinations and bias
Potential hallucinations and bias in model outputs can shape casting and introduce stereotyped background characters. This risk underscores the importance of high-quality models, bias testing, and human review before AI-generated outputs reach the consumer. “Bias shows up in HR (for example, casting workflows) and marketing (such as only generating images of white men),” said a leader in AI ethics. “You can absolutely see it in creative pipelines when tools influence representation.”
Our research finds that AI is already demonstrating potential to reshape pre- and post-production. Much greater change could be likely as the technology evolves, but what form that future takes remains uncertain. Given this uncertainty, media executives should prepare to respond to a range of possible ways AI could have an impact on their business. They could consider, for example, experimenting with fair and appropriate scaling AI in production and establishing clear markers and milestones to guide their strategy amid the uncertainty. Ultimately, those who understand the potential implications, prepare for the risks, and begin to rewire their organizations for AI where needed are likely to help define film and TV in the decade ahead.
