AI-generated performers have moved from the laboratory to cinema trailers and casting announcements, prompting fresh debate about the creative and ethical limits of the technology. A recently released preview for the western As Deep as the Grave features a digitally recreated younger Val Kilmer — produced with the cooperation of his estate — while the synthetic persona known as Tilly Norwood is scheduled to appear in the feature film Misaligned.
Surface realism, deeper doubts
The visual effect in the Kilmer trailer is striking enough to provoke strong reactions, but industry practitioners caution the result still falls short of a convincing human performance. Tim Webber, chief creative officer at UK visual effects company Framestore, said current models are trained on surface data and mimic external appearance rather than the interiority of a performance.
“AI models are trained on the surface of things – the pixels of internet images, if they’re image models. To get to the level where it’s a human performance, where there’s genuine emotion involved, is a long way further down.”
Those limitations are visible in the Kilmer likeness: observers described the portrayal as unnerving and mechanically stiff, with a voice and the actor's well-known presence notably absent. The Kilmer estate authorized the use of his likeness, noting his interest in technological experiments that expand storytelling, but even supporters concede the result can be unsettling.
Public reaction and artistic concerns
Reactions to synthetic performers have been intense. When Tilly Norwood was introduced in a short comedic sketch, several actors and members of the public responded negatively. One high-profile reaction came from Emily Blunt, who described the synthetic figure as “really, really scary,” reflecting a broader unease about the prospect of artificially generated faces and voices occupying creative roles.
Beyond immediate aesthetic discomfort, the emergence of AI actors raises practical and ethical questions:
- Who controls consent and rights when a digital likeness is used?
- Can AI capture the subtleties of human expression and improvisation that define many performances?
- What impact will synthetic performers have on employment across acting, casting and post-production?
Industry leaders, unions and studios have yet to converge on answers. The technology's proponents argue that AI tools can expand creative possibilities, enabling projects that would otherwise be impossible. Critics warn of a technology-driven agenda that risks imposing synthetic entertainment on audiences and displacing workers.
Where the technology stands
Technical development is rapid, and firms are experimenting with different approaches to synthesise faces, bodies and vocal patterns. But the distinction highlighted by Webber — that current systems reproduce surface features rather than the lived, interior dynamics actors bring to roles — remains central to professional scepticism.
| Aspect | Human actor | AI-generated performer |
|---|---|---|
| Source of performance | Internal experience, interpretation | Model trained on images and recordings |
| Emotional authenticity | High (subjective, variable) | Currently limited to surface mimicry |
| Consent and rights | Established industry practice | Legally and ethically contested |
The table summarises distinctions already evident in the debate; it does not attempt to predict technical breakthroughs but shows why many practitioners remain cautious.
Industry implications for the UK
For Britain’s screen sector — home to major VFX houses and the global production pipeline — the growth of AI-generated performers presents both opportunity and risk. Firms that can supply convincing synthetic characters may win business, but studios and workers will press for clear rules governing likeness rights, credit, pay and safety. The debate is likely to extend into negotiations with performers’ unions and regulatory scrutiny over intellectual property and data use.
At present the evidence suggests AI can create compelling imagery and provoke discussion about storytelling possibilities, but the claim that it will imminently replace human actors is not supported by the quality of current work or by expert commentary. The technology is improving fast, yet for the moment it is imitator rather than originator: able to reproduce faces and gestures, but not the lived authenticity that underpins many memorable performances.
How the industry responds — through codes of practice, contractual safeguards and technical standards — will determine whether AI becomes an imaginative tool for filmmakers or a disruptive force that undermines livelihoods and erodes trust in what audiences see on screen.
