AI is already starting to be deployed in some areas of the film and TV production course of, although the potential magnitude of its long-term affect remains to be coming into focus. Our analysis and discussions with studio executives, producers, and expertise leaders recommend that uncertainty round AI extends past whether or not and the way it will change production to how these modifications manifest all through the content material and distribution ecosystem. While the expertise’s limits, adoption trajectory, and potential scale of affect are but to be decided, historic technological shifts and early use circumstances recommend AI could, over time, materially alter the industry’s construction and revenue swimming pools.
As a end result, {industry} leaders face sensible questions on near-term working decisions and strategic questions on what AI could mean for their companies long term. Based on our current expertise, analysis, and discussions, AI’s increasing capabilities have prompted some leaders to start to reevaluate their enterprise methods whereas recognizing they need to additionally handle looming considerations about labor impacts, potential dangers, and the nature of creativity. As Sean Bailey, an {industry} veteran and founding father of B5 Studios described the problem to us, AI could characterize “a more significant platform shift than we have ever seen before in our industry.”
To perceive how AI could affect the total video content material {industry}, we interviewed over 20 media leaders, together with studio and production executives, expertise brokers, AI innovators, and teachers; drew learnings from our work with video content material firms; and analyzed broader {industry} information and the historical past of expertise improvements in content material production and distribution (see sidebar, “About the research”).
The rise in AI comes at a second when video {industry} gamers are already below immense stress. Consumer consideration is fragmented amid an abundance of content material and finite viewing time, and consideration is shifting away from conventional channels to streaming platforms and user-generated content material (UGC). In the United States, for occasion, every day viewing hours spent on linear TV declined by 4 % CAGR from 2022 to 2024, whereas streaming grew by 13 % and social video platforms grew by 14 % (Exhibit 1). Consumers are additionally altering how they watch, consuming video on cell units and more and more utilizing their TVs to go looking for and view on-line movies, together with user-generated content material.
At the similar time, funding in content material is leveling off. In the United States, which represents over half of worldwide spend, unique content material spend is projected to say no by 2 % per yr as patrons flip to sports activities rights and licensed programming, which might entice outsized audiences or price much less (Exhibit 2). These projections don’t embody the potential affect of AI, which introduces important uncertainty round content material spending.
Amid these shifts in provide and demand, leaders interviewed indicated AI has the potential to affect many production processes, in addition to back-office operations. New instruments and early experimentation are already demonstrating single-digit productiveness enchancment potential in some use circumstances, whereas additionally elevating essential questions on mental property (IP), authenticity, and the future of artistic work.
While the longer-term implications are usually not but clear, this analysis seeks to determine some potential {industry} outcomes, beginning with already-emerging impacts on production workflows. It additionally explores how AI could change what content material is created and who creates and distributes it over the long term below various adoption eventualities.
The attainable outcomes mentioned embody the democratization of high-end content material creation, accelerating the shift of client consideration from professionally produced content material to UGC platforms and smaller studios, in addition to fully new types of content material and distribution, together with extra immersive, customized, or participatory leisure. Our report additionally identifies the potential redistribution of financial worth and the potential for a web improve in complete content material provide and demand if AI’s affect is just like previous disruptions from the rise of recent production and distribution applied sciences. This contains potential impacts on production workflows, influencing roughly 20 % of unique content material spend in the subsequent 5 years and redistributing as much as $60 billion of annual income inside 5 years of mass adoption of AI use surpassing that of incumbent expertise.
Many in the {industry} have raised considerations about how AI adoption could have an effect on these potential outcomes and different attainable eventualities, together with the affect of AI on creativity and the already-declining variety of leisure jobs, whereas others query whether or not sure makes use of of AI ought to be permitted in any respect in artistic production. Current adoption and affect is uneven, and its future trajectory is unsure. As a number of executives made clear, AI-generated output will not be but at a high quality stage to drive significant disruption, with content material that in lots of circumstances doesn’t meet premium production requirements. While this will likely change, leaders interviewed famous that there are limits to how deeply AI will disrupt end-to-end content material creation and the artwork of storytelling in established TV and film codecs.
This report is restricted to exploring the enterprise, not artistic, implications of AI, and it doesn’t assess the decisions that could come up as the affect turns into clearer (for instance, how greatest to make the most of the productiveness enhancements AI could allow or the demand for new artistic roles and expertise even when present ones are affected).
In addition, a number of studios have challenged relationships with foundational fashions and publicly obtainable text-to-video instruments that they declare have been illegally skilled on the studios’ IP. And there are important unresolved points and debates round authorship and rights possession, particularly from members of the artistic group.
With all these forces at work, many essential uncertainties stay about the affect of AI on content material production, innovation, and the longer-term distribution of financial worth throughout the ecosystem. In the absence of clear solutions, the insights that observe are supposed to supply {industry} leaders with a basis for assessing what it might mean for their companies operationally and strategically, what markers to observe, and what inquiries to ask as AI capabilities proceed to develop.
AI is already displaying potential to reshape some core production actions
Leaders we interviewed report experimenting throughout choose production processes and discovering potential for 5 to 10 % productiveness will increase in particular use circumstances this yr. Much of this preliminary worth comes from trials throughout improvement and pre-production workflows, with leaders telling us they count on to develop to bodily production and post-production over the subsequent 5 years.
Development and pre-production at the forefront
Development and pre-production will be significantly time-consuming, particularly when speaking artistic imaginative and prescient. Leaders famous that, whereas it’s nonetheless early, extra developed use circumstances exist on this space than in different production phases, partially as a result of they require much less technical complexity and pose decrease adoption danger, making it a logical start line. Indeed, our analysis finds that these early adopters are already realizing mid-single-digit productivity-related will increase throughout the pre-production course of, with extra in sure genres or workflows. “People are now coming to pitches with clearer visual directions and pre-production steps [using AI],” mentioned a former govt at a streaming firm. Leaders interviewed famous longer-term potential for pre-production modifications to have an effect on later phases, reminiscent of bodily production and post-production. This could embody a mindset shift from the norm of “fix it in post” to “fix it in pre,” based on Hannah Elsakr, Adobe’s vice chairman of gen AI new ventures. Similarly, use circumstances together with A/B testing of story beats and analytics-led script breakdowns that determine production components, reminiscent of location, props, and shot lists, could enable groups to be extra focused and shoot sooner when on set.
Room for development in bodily production
Physical production, together with set building, principal images, and reshoots, represents the majority of production exercise. Despite some examples of AI use in bodily production, reminiscent of Netflix’s first unique sequence with AI-generated ultimate footage, early adoption lags that of pre- and post-production. Adoption obstacles cited by {industry} executives embody limitations of present instruments; protections for artistic expertise throughout IP, contracts, and labor; and considerations round client preferences. “Creative integrity is so important,” mentioned Kevin Lingley, govt vice chairman 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 recognized potential for two long-term shifts. First, on-location and capturing processes could change. “Think about car chases, where you historically had to lock down [parts of] a city for two months,” mentioned a former govt of main film studio. “Instead, we could re-create locations virtually.”
Second, bodily production could be accelerated, each by shifting work into pre-production and by lowering reshoots. “Shorter production cycles are a huge advantage to maintain momentum in a hit-driven business,” defined Jan Lacher, senior vice chairman of content material and enterprise improvement, RTL Group.
Usage rising in post-production
Post-production work is commonly outsourced and will be the second most intensive step, significantly for VFX-heavy sci-fi and blockbuster films. Producers famous that AI is already used to dub and localize content material and to speed up footage clipping and video library filtering, a development illustrated by Moments Lab’s partnership with Banijay. Many highlighted that further makes use of are nonetheless to return. For instance, a former govt at a serious media firm projected that after AI is adopted, it could speed up present processes and ultimately outline new AI-native workflows in animation and VFX tasks. Once instruments attain what RTL’s Jan Lacher referred to as “professional-grade resolution and consistency,” put up schedules could shorten considerably, based on leaders, and AI could begin to mix post-production into the pre-production course of. “That is revolutionary in terms of the way we make content,” mentioned Adrienne Lahens, CEO and cofounder of Infinite Studios.
What previous improvements could point out about AI’s potential {industry} affect
While there are early indicators of affect as organizations experiment with alternatives in content material creation, the longer-term implications for the total {industry} stay much more unsure, with a number of attainable outcomes.
To perceive how the {industry} may evolve, relying on the extent of AI adoption, our analysis explored what ten of the most vital previous content material production and distribution expertise improvements may educate us about AI’s potential (Exhibit 3). Four major classes emerge from this evaluation:
- New production expertise drove important change in {industry} economics, redirecting some spend to tech distributors and passing most of the worth to distributors. The introduction of digital cinematography illustrates this sample. It marked a shift from film inventory and processing prices to digital production and enabled sooner, extra versatile workflows on-set. Some of the worth delivered was captured by the distributors who supplied this new expertise, however in lots of circumstances distributors gained the majority of the worth at stake from production expertise shifts by way of greater revenue margins. An analogous redistribution of worth could happen with AI because of the structural fragmentation of producers versus focus of distributors, with seven patrons making up 84 % of US content material spend.
- Historically, most worth from production improvements accrued to massive incumbents with the means to take a position, whereas customers have benefited in a number of methods. For instance, the first cameras have been accessible solely to massive skilled studios, and computer-generated imagery (CGI) expanded extremely costly effects-driven genres like sci-fi and fantasy. CGI improved the visible constancy of movies, elevating the bar for premium productions, and common Hollywood blockbuster budgets have elevated by 30 % over the previous twenty years, with CGI shaping the highest-budget movies. Home video cameras, PCs, and cell phones have put video seize and enhancing instruments in the arms of customers, however AI could also be the first expertise with the potential to allow creation of content material at greater production values exterior of the massive skilled studios if instruments enhance sufficiently.
- Most distribution improvements expanded client alternative and usually elevated total content material provide and demand. Television opened up extra viewing events and time to observe and required extra video content material to fill the obtainable capability. The VCR and DVD made it attainable to each seize and view movies on demand at house, creating extra management and extra alternative, together with the rise of “direct-to-video” movies. UGC platforms enabled entry to an abundance of free-to-watch, on-demand content material throughout area of interest genres and codecs. Consumers now watch hours of user-generated content material per day (see Exhibit 1).
- New applied sciences are sometimes adopted in surprising methods, giving rise to new content material codecs and distribution fashions. For instance, early filmmakers handled the digital camera as a software for recording stage performances, somewhat than the foundation for new types of storytelling that it turned. Television programming first targeted on transmission of reside occasions and training lectures but went on to help new episodic content material codecs. Leaders famous that the use of computer systems in the animation style was first thought of “niche” till Pixar developed into one among the largest animation studios. Most just lately, mobile-phone cameras performed a major, if not main, position in the rise of short-form content material codecs and the open distribution platforms which might be main locations for customers and advertisers.
Three potential methods {industry} dynamics could evolve
Together, these classes from previous improvements and our broader analysis recommend three potential industry-wide outcomes from the scaling of AI: scaling of modifications to present production workflows, wide-scale democratization of professional-grade content material creation, and the creation of recent content material codecs and distribution channels. These outcomes, which aren’t supposed to be exhaustive or mutually unique, are offered so as of their chance of occurring, primarily based on our skilled interviews and evaluation. Whether and how they finally play out (or different outcomes take form) will depend upon the extent to which high-end AI production instruments develop into democratized (and who captures the ensuing worth), creatives undertake these instruments throughout workflow steps, client viewing preferences change, new codecs develop, and new distributors and aggregators emerge.
1. Scaling of modifications to present production workflows
Leaders we interviewed famous that it’s doubtless AI will more and more affect production workflows, given the early indicators 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,” mentioned B5 Studios’ Sean Bailey. Our evaluation signifies that roughly $10 billion of forecast US unique content material spend in 2030 could be addressable by some type of AI.
While adoption stays restricted as we speak, leaders interviewed famous a lag between technological improvement and industry-wide uptake, as seen in the historic evaluation. CGI offers a transparent precedent. As highlighted by Bryn Mooser, cofounder of Asteria, Jurassic Park’s landmark results “changed the entire industry,” finally growing film budgets, though Lucasfilm’s Industrial Light & Magic had been based practically twenty years earlier. However, interviewees mentioned {industry} gamers might want to tackle the many considerations about the expertise’s use in creative and artistic endeavors, reminiscent of the way it impacts creativity or authorship.
Interviews and previous historic classes recommend distributors are positioned to seize the majority of worth delivered from AI-driven will increase in workflow velocity and capability. This is pushed by structural market dynamics, together with a crowded producer market, consolidating purchaser panorama, and transparency of production budgets. Producers who make investments in new applied sciences, adapt their working fashions, and have or develop robust IP are additionally properly positioned to seize a portion of this worth.
How a lot financial worth expertise distributors could seize stays unclear. If broader video instruments observe the trajectory of enormous language fashions, with intense competitors between distributors and open-source fashions carefully trailing best-in-class choices, then worth could largely cross by way of to distributors and producers as instruments develop into democratized and cheap. However, if high fashions meaningfully diverge from others, tech distributors could modify their pricing, capturing a larger share of worth. Leading mannequin builders and high IP homeowners could additionally kind unique partnerships for professional-grade video fashions, constructing a closed-loop ecosystem, whereas conventional production service suppliers, reminiscent of movement seize suppliers or particular results (SFX) specialists, could doubtless face mounting stress as routine duties develop into automated.
2. Wide-scale democratization of professional-grade content material creation
AI could additionally allow smaller studios and artistic entrepreneurs to compete extra immediately with massive studios, with the potential to extend complete content material provide and open up new alternatives for the broader artistic group. The analysis signifies this consequence is much less sure than the anticipated affect throughout present production processes. It could rely partially on whether or not the smaller creators apply AI to ship extra high-end content material somewhat than merely to extend the quantity of content material they produce. Already, early use of AI instruments has led to the improvement of what many confer with as AI “slop.” Interviewees additionally acknowledged that incumbents nonetheless have distribution benefits. “In a world with even more content, IP owners will have a relatively higher chance of success,” mentioned Michael Porter, an {industry} skilled. “Their brands are more likely to cut through the noise in a world saturated with content.”
If widespread democratization of professional-grade content material does happen, it could put further monetary stress on conventional producers and distributors. Large incumbents would want to rewire long-standing workflows and working fashions as this occurs, whereas newer studios could adapt extra shortly. Entrepreneurial AI-enabled creators and artists could inform their tales in new methods and open distribution platforms. “There are an ever-increasing number of entry points for content,” mentioned Matthew Wilson, chief authorized officer, Fremantle. “AI only adds to our ability to play across new distribution channels.” The mixture of finite client consideration and extra content material throughout extra distribution factors could have a major affect on the present distribution panorama. A doubtlessly analogous disruption adopted the rise of broadcast TV, which contributed to a 38 % drop in the variety of cinemas between 1930 and 1957 in the United States.
Even a modest shift in viewer conduct could have a significant affect on {industry} revenue swimming pools. If, for instance, present open video distribution platforms captured an incremental 5 % of TV and film viewing hours in the United States, there can be a $13.2 billion decline in US TV and film distribution revenues. This loss would solely be partially offset by a $7.5 billion improve in open-platform revenues. The roughly $5.7 billion lower in web income on this instance displays a shift in viewership to open platforms with decrease monetization per hour.
3. The creation of recent content material codecs and distribution channels
To the extent that AI offers rise to new content material codecs and distribution platforms, not simply shifting distribution throughout present platforms, it could considerably redistribute financial worth swimming pools amongst {industry} individuals. The shifts from stage performs to cinema, linear to streaming, and long-form to short-form every contracted incumbent income by a median of 35 % in the 5 years after the expertise was broadly adopted, with customers receiving a wider collection of cheaper content material. While AI remains to be early in its adoption curve, making use of this historic sample to forecast income signifies that round $60 billion of income could be redistributed inside 5 years of reaching mass adoption, if AI use surpasses that of incumbent expertise.
Specifically, what these new codecs could appear like is troublesome to evaluate, given the nature of the artistic course of. However, many executives mentioned such a shift is feasible, primarily based on previous 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,” mentioned RTL’s Jan Lacher.
Dani Van de Sande, founder and CEO of Artist and the Machine, famous extra transformative shifts could come as AI strikes towards world fashions, methods that don’t simply generate remoted property however keep an inside understanding of characters, environments, guidelines, and trigger and impact 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 mentioned.
From right here, quite a few structural shifts could come up if historic patterns maintain.
Fundamentally new fashions could emerge, together with built-in platforms that mix creation and distribution in a single atmosphere, just like how TikTookay and CapCut allow AI-supported creation and near-zero-cost international attain. “Creators will move upstream as AI puts cinematic quality production tools in the hands of people who never had access to traditional Hollywood pipelines,” mentioned 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 instance is DreamFlare, a hybrid creation and distribution platform the place creators can publish episodic, AI-enhanced visible tales on to audiences, who can vote on which ideas are developed into full exhibits. This could allow new storytelling codecs past as we speak’s conventions.
AI could additional reshape viewers conduct and expertise. Streaming platforms have already moved audiences away from a monoculture, partially by studying person preferences and dramatically increasing the provide of content material and vary of particular person alternative and management, a transition that additional AI-driven hyper-personalization could speed up. “One [shift] is that the user is driving the content that they want to see,” mentioned 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 expertise reshapes the production course of, leaders emphasize the worth of artistic expertise and style, citing premium content material anchored in human-led storytelling as a defining marker of high 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?” mentioned Lori McCreary, producer and CEO of Revelations Entertainment and a board member of the Producers Guild of America.
Those constructing video, picture, and textual content fashions emphasised that human-led storytelling could additionally affect how the expertise develops. For instance, Runway’s Michelle Kwon envisions an integration amongst software distributors into unified artistic platforms. “There won’t be LLM companies and video generation companies,” she mentioned. “The whole industry is moving toward world models.”
Ethical and danger concerns raised in our interviews
However the {industry} evolves, AI’s increasing position in content material creation brings dangers, lots of that are already right here. Across our interviews, three considerations emerged most persistently: expertise and artistic implications, IP and different rights infringement, and potential hallucinations or bias in mannequin outputs. Leaders interviewed pointed to the essential position of regulatory and moral frameworks, together with measures reminiscent of coaching fashions on IP-safe content material, standardizing AI use compensation, and guaranteeing creator consent with significant management over digital likeness.
Talent and artistic implications
Creative erosion and associated expertise implications heart on whether or not AI-generated or AI-modified work can replicate creative intent and precisely painting lived experiences with out distortion, in addition to on AI’s final affect on leisure jobs. Guilds and unions have negotiated for language that protects artists. Recent occasions highlighting the sensitivity of those points embody the SAG AFTRA (Screen Actors Guild–American Federation of Television and Radio Artists) and WGA (Writers Guild of America) strike and debates over the film The Brutalist, which used AI dubbing and voice modification to create extra sensible Hungarian accents. The rise of digital likenesses provides complexity for expertise, producers, and distributors attempting to make sure truthful compensation and shield in opposition to deepfakes.
Talent companies, one other key stakeholder, could create new service traces serving to shoppers handle, monetize, and shield digital likeness, voice, and IP rights at scale, even when modifications to production workflows disrupt their conventional companies mannequin. “We are focused on setting perimeters and protections for talent, IP owners, and rights holders,” mentioned Alexandra Shannon, an govt at Creative Artists Agency (CAA). “Premium, authentic, human-led content will be even more valuable if the overall supply of content increases.”
IP and different rights infringement
Concerns relating to IP and different rights infringement have led to lawsuits over coaching fashions which have used present mental property. These points are prompting {industry} leaders to think about what Hannah Elaskr known as the “nutrition label” for model-training information. They are additionally giving rise to new partnerships between studios and AI firms to develop IP-protected proprietary fashions which might be skilled solely on licensed information. “We have new types of legalities to worry about as producers,” mentioned 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 danger of hallucinations and bias
Potential hallucinations and bias in mannequin outputs can form casting and introduce stereotyped background characters. This danger underscores the significance of high-quality fashions, bias testing, and human overview earlier than AI-generated outputs attain the client. “Bias shows up in HR (for example, casting workflows) and marketing (such as only generating images of white men),” mentioned a pacesetter in AI ethics. “You can absolutely see it in creative pipelines when tools influence representation.”
Our analysis finds that AI is already demonstrating potential to reshape pre- and post-production. Much larger change could be doubtless as the expertise evolves, however what kind that future takes stays unsure. Given this uncertainty, media executives ought to put together to answer a spread of attainable methods AI could have an effect on their enterprise. They could contemplate, for instance, experimenting with truthful and applicable scaling AI in production and establishing clear markers and milestones to information their technique amid the uncertainty. Ultimately, those that perceive the potential implications, put together for the dangers, and start to rewire their organizations for AI the place wanted are doubtless to assist outline film and TV in the decade forward.
