SAG‑AFTRA, the ‘Tilly Tax,’ and the Labor Moment That Will Define Work in an AI Era
For generations, the labor of actors has been both literal and intangible: a body, a voice, an epoch of presence that cannot be separated from the human moment it inhabits. Today, that presence is being reframed as data. Face scans, motion capture, voice prints and performance datasets transform an actor into a stream of signals that can, with growing sophistication, be stitched back together without the human who first gave them life. That shift is not merely technical. It is a workplace crisis, a question of control over labor, ownership of likeness, and the distribution of value created from human creativity.
The bargaining table where technology meets livelihood
At the center of this storm is a negotiation unfolding between SAG‑AFTRA and the studios: a compact over how artificial intelligence can be used to replicate, extend or replace actors in film and media. The union is pushing proposals that aim to limit the use of synthetic likenesses without express consent and to ensure performers are compensated when AI systems draw on their work. Among the ideas on the table is a proposal that has captured headlines and imaginations alike: a so‑called Tilly tax, a levy on productions that use AI replicas or synthesized performances.
Why a tax? Not as a punitive flourish, but as an instrument of agency. The Tilly tax is designed to do more than raise money. It is a mechanism to tilt incentives: to make studios think twice about substituting living performers with cheaper, easier digital copies. It would put a price on replication, finance retraining and transition funds, and create a visible accounting of the degree to which technology is replacing human craft.
Beyond symbolism: the practical logic of a levy
A levy like the Tilly tax is rooted in three practical goals. First, deterrence. If studios can deploy AI replicas at near zero marginal cost, the economic pressure to replace people becomes enormous. A fee raises the marginal cost, preserving space for negotiation and meaningful consent.
Second, compensation. When a studio uses an AI likeness, the resulting product profits from a likeness that sprang from a performer’s labor. A dedicated fund can channel fees into residuals, payments to performers whose images and voices are used, and support for those whose work is disrupted.
Third, adaptation. Technology-driven displacement is not a one‑time shock but a continuing transition. The Tilly tax envisions money for retraining, education, and pathways into new creative work where actors remain central—performance capture direction, voice coaching for AI interaction design, and creative oversight roles that ensure human nuance in storytelling.
What this fight teaches the wider world of work
The argument is not just about Hollywood; it is a field test of a broader labor principle. Across industries, faster and cheaper automation threatens to hollow out the meaning of skilled work. The response emerging in film is instructive: unions asserting that technological change must be governed in ways that preserve workers dignity, agency, and share of value.
For the broader Work community, the bargaining over AI in film is a playbook. It demonstrates how to translate abstract fears about automation into contractual language: clear definitions of permitted and forbidden uses, mandatory transparency about datasets and model training, audit rights, revenue sharing for AI‑derived outputs, and enforceable penalties for misuse. It confronts a core question of our time: who gets to profit from digitized human labor, and how do we prevent the concentration of returns in the hands of the few who own the models and platforms?
Defining consent and control
Consent is the pivot. At the moment, consent in entertainment has often been episodic: a release form for a shoot or a clause in a contract. But if an actor’s face or voice can now be recombined to produce performances they never gave, consent must be continuous and contractually protected. That means explicit opt‑ins for the creation and commercial use of synthetic likenesses, renegotiable terms when a replica is used beyond the original context, and the right to revoke permission in certain circumstances.
Control is equally crucial. That can take many forms: a prohibition on the use of replicas to substitute for live performers, limits on the contexts where synthetic likenesses are allowed, or technical safeguards like watermarking and embedded provenance metadata so downstream viewers and creators can know when a performance is synthetic.
Mechanisms for accountability and verification
Promises mean little without enforcement. If the industry adopts rules that restrict AI use but lacks a practical way to check compliance, the protections become performative. This is where contractual audit rights and industry registries become powerful tools. Producers could be required to maintain tamper‑evident logs of when and how a synthetic likeness is generated and used. Independent audits, funded by modest levies, would verify compliance and, where violations occur, trigger penalties that dwarf any short‑term savings from skirting the rules.
Technology also offers solutions. Digital signatures, cryptographic watermarks and provenance chains can make it straightforward to identify AI‑generated footage. A universal label applied to synthetic performances would not eliminate misuse, but it would make transparency the default — and that matters both for protecting performers and for informing audiences.
Rethinking value and craft
At its heart, this negotiation is a debate about what we value in creative labor. Is performance merely a set of extractable assets — a face, a voice, a motion dataset — or is it an embodied craft, embedded in context, improvisation, and the chemistry between people on a set? A digital copy can mimic surface features, but it struggles to mimic presence: the microtuned timing, the lived experience that colors a choice. Those intangibles have economic value, and labor strategies like the Tilly tax are efforts to keep that value recognized and compensated.
Moreover, actors are not the only professionals whose livelihoods depend on being recognized as more than data. Camera operators, grips, dialect coaches, editors and others form an ecosystem where craft and collaboration create worth. Policies that protect performers from unconsented replication rarely fail to protect the broader ecosystem.
Designing a future where technology amplifies rather than erases
There is a middle path between romantic techno‑suspicion and uncritical acceptance. Technology can amplify creative possibilities: enabling new forms of storytelling, expanding access to performance tools, and creating new revenue streams. The challenge is to build a framework where those opportunities do not come at the expense of workers’ bargaining power.
Practically, that framework looks like this: clear contractual rules, shared registries of synthetic assets, a tiered levy system like the Tilly tax that scales with the degree and profitability of synthetic substitution, dedicated funds for retraining and transition, and enforceable transparency obligations backed by independent audits. Combined, these elements channel the benefits of AI into creative expansion while ensuring that human contributors remain central and fairly compensated.
Why this matters to every workplace
Actors are the canary in the coal mine for a new era of data‑driven labor. When a performance can be synthesized, labor negotiations cannot be reduced to hourly rates alone. They must also be about data rights, provenance, consent and the institutional measures that distribute value and protect dignity. How this bargaining ends will send signals across sectors: whether workers can insist that automation be rolled out under rules that protect human agency, or whether capital can automate first and ask forgiveness later.
For managers, policy makers and workers outside entertainment, there is a lesson here: technological change is not inevitable in how it affects people. It is shaped by policy, bargaining strength and the design choices we make now. If we want a future where technology augments and uplifts labor, not replaces and marginalizes it, this is the kind of fight we must be ready to wage in every industry: from manufacturing lines to legal offices, from customer service centers to laboratories deploying synthetic biological tools.
A call to solidarity and imagination
The Tilly tax and related proposals are not anti‑innovation. They are a plea for imagination: an insistence that innovation be governed in ways that safeguard work, creativity and dignity. Bargaining over AI use in film is, at its best, an exercise in democratic stewardship — deciding how new tools will be integrated into social and economic life.
That stewardship requires people across the workforce to pay attention, to demand clarity about how data drawn from human labor will be used, and to push for mechanisms that share the returns of automation more broadly. If artists can insist that their likenesses remain their own and that any use of synthetic replicas be governed by consent, transparency and fair compensation, other workers can press for similar protections around the models and datasets that increasingly define their jobs.
Conclusion: making policy that honors human work
Technology will continue to reshape how work is done. The path we choose now — whether to allow unbridled substitution or to craft rules that align innovation with human values — will determine whether the next generation inherits a labor market that respects skill, protects identity and shares gains. SAG‑AFTRA’s bargaining, including the push for measures like a Tilly tax, is not a niche cultural fight. It is a labor manifesto for the AI age: a call to hold technology accountable to the people whose labor made it possible.
For the Work community, the lesson is clear. Protecting livelihoods in the age of AI requires more than new laws. It requires imagination in bargaining, industry standards that prioritize transparency and fairness, and a social contract that recognizes that data derived from human work should not be an extractive resource but a shared asset. This is the moment to insist that technological progress be measured not only by the speed of innovation, but by the measure of lives it sustains and uplifts.
