Welcome to the Synthetic Century
AI is learning to fake everything—actors, products, and even places. The next generation of stars won’t be discovered; they’ll be designed, trained, and corporately owned.
A still of the computer-generated Tilly Norwood. (Xicoia)
We’ve entered the age of the synthetic — a world where the “real” is no longer the default, just the reference material. AI video tools, digital humans, and generative design engines are beginning to outproduce their analog counterparts. Products, places, and people are being replicated in code, rendered in high definition, and distributed at scale.
The creative industry tells itself this is about imagination. In truth, it’s about substitution. The new frontier of media isn’t about what’s real—it’s about what’s profitable to fake.
When the AI talent studio Particle6 introduced Tilly Norwood, the first fully synthetic actress, founder Eline van der Velden called her “a creative work,” not a replacement. It was the right thing to say. But it’s hard to ignore that Tilly costs nothing to feed, transport, or unionize—and can perform indefinitely.
This looks much better than you may think.
Her debut short, AI Commissioner, used 16 AI-generated characters built from ten different models. The result looked like an uncanny hybrid of a Pixar simulation and a corporate explainer video. Critics mocked it. Studios noticed anyway. Because even if the performances weren’t perfect, the economics were.
Synthetic actors promise up to 90 percent cost reductions, according to Particle6. That’s a business case, not a creative argument. The same math now drives the rise of generative video platforms like Sora 2, Runway, and Synthesia—tools that can generate film-quality scenes from a text prompt. The frontier isn’t performance. It’s production.
The $1 Million Lesson
The shift came into focus this spring when Jon Gray, President and COO of Blackstone, showed two commercials at the firm’s annual CIO Symposium. The first was shot conventionally—on location, with real actors, a professional crew, and full post-production. Cost: $1 million.
The second was AI-generated. Same concept. Same pacing. Two Blackstone staffers built it in an afternoon using a generative video platform. The cost was effectively zero.
The crowd of investors didn’t see an experiment in filmmaking. They saw a new cost structure. Gray didn’t have to say the quiet part out loud: this is the model of the future—creative output measured not in artistry, but efficiency. See for yourself.
From Real to Replaceable
Synthetic media won’t stop at actors. Once you can generate people convincingly, you can generate everything else. Locations, products, and even entire brands can be fabricated in silico, then deployed instantly.
A film that once required shooting in Venice can now generate “Venice” in a prompt. A fashion brand can produce a campaign without garments, models, or photographers. A travel company can market destinations no one has ever visited—and maybe never will.
Synthetic content isn’t about mimicking the world; it’s about replacing the need to leave the studio at all. The world becomes a dataset.
According to Precedence Research, the AI video market is projected to expand from $10.3 billion in 2025 to $156 billion by 2034, growing at a rate of more than 35 percent annually. Meanwhile, AI-image tools like Midjourney, Runway, and Pika are growing even faster, automating illustration, rendering, and visual design. The incentive is obvious: the synthetic world scales better than the real one.
Hollywood has always been a factory of illusion. Matte paintings became green screens; green screens became volumetric LED stages. Now, even that technology looks quaint. Why pay to rent a studio—or fly to Iceland—when AI can fabricate your backdrop with perfect lighting and continuity?
The location scout, the production designer, even the cinematographer—all are being absorbed into prompt engineering and simulation tools. The irony is that AI may finally deliver the perfect “cinematic realism” directors have chased for a century—by eliminating reality altogether.
The same dynamic is hitting commercial production. The $1M ad vs. AI demo Jon Gray presented at Blackstone was a proof-of-concept. The next step is integration: stock footage libraries, voiceover catalogs, and brand assets will merge into fully synthetic pipelines. Every variable—actor, product, place—will be adjustable on demand.
The Ownership Economy
The current generation of film and television stars may find ways to monetize this shift. Some will license their likenesses, voices, and gestures for digital replication. Their agents will negotiate “synthetic appearance rights,” ensuring residuals every time their digital doubles appear on screen.
But that model ends with them. The next generation of stars may be entirely artificial—built from scratch, trained on aggregated facial data, and corporately owned.
In that world, the notion of celebrity changes. A human actor represents autonomy, negotiation, and the unpredictability of fame. A synthetic one represents pure control. Studios and brands will no longer have to deal with scheduling conflicts, public scandals, or contract renegotiations. The face of your franchise becomes software IP.
It’s not hard to imagine an entire cinematic universe—actors, locations, props, score—existing as an internal asset on a studio’s balance sheet. The question of authorship becomes irrelevant when everything is owned.
In The End of the Blockbuster, Scott Galloway argues that the modern entertainment machine collapsed under its own weight:
“The Marvel movie’s staff—from visual-effects artists and animators to costume designers and location scouts—is bigger than the entire workforce of Lyft or Reddit.”
I mean, just ponder the inputs here:
Galloway cites David Ellison, founder of Skydance Media, who calls technology a “powerful multiplier” of creativity, not a replacement for it. That may be true in theory, but a multiplier always amplifies what it’s given—and in today’s Hollywood, the dominant motive is cost discipline.
The Animation Guild’s 2024 report predicts that 118,500 jobs—roughly 21 percent of the U.S. film, TV, and animation workforce—are vulnerable to AI automation. In California alone, 62,000 positions could disappear within three years. Visual effects, post-production, and location departments are already contracting as studios experiment with generative pipelines.
Even among those who remain, optimism is scarce. A 2025 survey of creatives found that only 19 percent use AI for story development or brainstorming. Most use it for production efficiency—editing, rendering, or asset management.
This is the inevitable direction of the creative economy: from authorship to automation, from inspiration to iteration. Once the infrastructure for synthetic production exists, reality itself becomes a premium add-on—a luxury.
Think of it as the “organic food” paradox for culture: the real will still exist, but it will cost more and be consumed for status. Real actors, real locations, and authentic performances will become niche—reserved for prestige projects, theater, and high-end campaigns. Everything else will be synthetic by default.
For a while, audiences may claim to prefer authenticity. But history suggests they’ll adapt. They already watch deepfake Tom Cruise on TikTok, CGI influencers on Instagram, and AI-generated singers on Spotify. The uncanny becomes familiar surprisingly fast.
The most profound loss won’t be economic. It will be philosophical. Creative industries have always been built on the tension between reality and illusion—between what’s captured and what’s crafted. Once that distinction collapses, creativity becomes pure simulation, endlessly reproducible and detached from human experience.
The coming generation of artists may not fight this shift; they’ll be born into it. I do think a new generation of artists will emerge, one that can embrace these tools and add a new level of humanity and creativity to the final products.
But before that happens, we will be wading in slop.
Climate Tech Atlas: Mapping the Path to Net-Zero Innovation
At Techonomy Impact, experts discussed the Climate Tech Atlas, a tool that maps the entire landscape of decarbonization technologies. They emphasized the need to direct talent, capital, and policy towards the solutions that matter most.
What I’m Reading
Google’s AI summaries have ‘devastating impact’ on news sites
I can tell you firsthand, this has affected all of my sites.Google’s new Web Guide search experiment organizes results with AI
A minor UI tweak with significant consequences—structured, topic-map browsing is how AI nudges search from “answers” to “research.”AI slop and fake reports are coming for your bug bounty programs
LLM-generated bug noise is the new DDoS: it burns triage cycles and erodes trust unless you upgrade incentives and filters.Apple Intelligence tries summarizing news again after botching BBC stories
Apple’s back with AI news summaries—this time with a warning label, which says as much about liability as it does about accuracy.Amazon buys Bee AI, a wearable that listens to everything you say
Ambient, always-on AI is marching from novelty to norm—and privacy will be retrofitted later unless users vote with their wrists.Google’s new AI feature lets you virtually try on clothes
Computer vision meets commerce: the best “storefront” increasingly starts with your camera roll, not a shopping mall.Samsung backs a video AI startup that can analyze thousands of hours of footage
Enterprise video is the next retrieval frontier—context across millions of minutes beats flashy short-clip demos.What America’s AI Action Plan Means for AI Governance
The policy tells you the playbook: centralize standards, speed infrastructure, and preempt the patchwork—regulation as industrial strategy.Last Week in AI #316 — ChatGPT Agent, AI IMO gold, $200M from DoD
Agents gaining computer control and models medaling at math signal the shift from chat to competent autonomy.OpenAI Seeks Additional Capital From Investors as Part of Its $40B Round
The capex curve is the moat—models, data centers, and talent now look more like national infrastructure than startup spend.