Role Definition
| Field | Value |
|---|---|
| Job Title | Stunt Performer |
| Seniority Level | Mid-level (3-7 years experience) |
| Primary Function | Executes high falls, fights, fire burns, vehicle work, wire sequences, and other dangerous physical action on film and television sets. Works under the stunt coordinator's direction, rehearsing and performing stunts that actors cannot or should not perform themselves. Maintains peak physical fitness and specialised skills (martial arts, precision driving, high-fall rigging, swimming, gymnastics). Also serves as body double for lead actors during action sequences. SAG-AFTRA covered, BLS SOC 27-2011 (Actors). |
| What This Role Is NOT | NOT a stunt coordinator (designs stunts, bears overall safety accountability — already assessed at 62.8 GREEN Stable). NOT an actor (primary skill is acting, not physical stunt execution). NOT a motion capture actor (performs in a capture volume for digital characters — already assessed at 46.3 YELLOW). NOT a circus performer (live audience, repertoire-based — already assessed at 60.0 GREEN Stable). |
| Typical Experience | 3-7 years. Typically entered through gymnastics, martial arts, or professional sports backgrounds. SAG-AFTRA registered. Specialised in one or more stunt disciplines (fights, falls, fire, vehicles, water, wire work). Strong physical conditioning and an injury-free track record essential for continued employment. |
Seniority note: Entry-level stunt performers (0-2 years, limited stunt credits, smaller gags) would score slightly lower due to less specialisation and weaker industry relationships. Senior performers (8+ years, department leads on tentpole films, borderline coordinator roles) would score similarly or higher — deeper specialisation and stronger personal brand moats.
Protective Principles + AI Growth Correlation
| Principle | Score (0-3) | Rationale |
|---|---|---|
| Embodied Physicality | 3 | Every working day involves performing dangerous physical action — falls from height, fight choreography with contact, fire burns, vehicle crashes, wire rigs — in unstructured, unpredictable film set environments. Each stunt site is unique: different terrain, weather, rigging points, equipment. Moravec's Paradox at maximum. |
| Deep Interpersonal Connection | 1 | Trust with the stunt coordinator is important — the performer entrusts their safety to the coordinator's design. Team coordination during multi-performer sequences requires non-verbal communication and physical trust. But the relationship is professional and task-specific, not the deep therapeutic or caregiving trust that scores higher. |
| Goal-Setting & Moral Judgment | 1 | Some judgment required — performers can and should refuse stunts they deem unsafe, and must assess their own physical readiness in real-time. But the creative direction and safety design come from the coordinator. The performer executes, not designs. |
| Protective Total | 5/9 | |
| AI Growth Correlation | 0 | AI adoption neither increases nor decreases demand for stunt performers. Demand is driven by audience appetite for action content and streaming platform investment, not AI trends. |
Quick screen result: Protective 5/9 + Correlation 0 — likely Yellow or Green Zone. Extreme physical danger is the primary moat. Proceed to quantify.
Task Decomposition (Agentic AI Scoring)
| Task | Time % | Score (1-5) | Weighted | Aug/Disp | Rationale |
|---|---|---|---|---|---|
| Physical stunt execution (fights, falls, fire, vehicles, wire work) | 40% | 1 | 0.40 | NOT INVOLVED | Performing dangerous physical action on active film sets — jumping from buildings, executing fight choreography with contact, performing fire burns in protective gel, driving vehicles at speed for camera. Unstructured environments, unique every time. No AI or robot can execute these stunts in the physical world on a live production set. Irreducible human work. |
| Rehearsal and physical preparation | 20% | 1 | 0.20 | NOT INVOLVED | Physically rehearsing stunts with the coordinator and team before cameras roll. Testing rigging, practicing timing and marks, working through fight choreography with partners. Requires the performer's actual body in the actual environment. No AI involvement. |
| Physical training and skill maintenance | 15% | 1 | 0.15 | NOT INVOLVED | Maintaining peak fitness, training in martial arts, gymnastics, precision driving, swimming, wire work. The body IS the tool. No AI can train a human body. |
| Body doubling / matching actor appearance | 10% | 3 | 0.30 | AUGMENTATION | Serving as a physical double for lead actors during action sequences. CGI face replacement and deepfake technology increasingly allow any performer's face to be replaced with the lead actor's in post-production, reducing the premium on physical resemblance. The performer still executes the stunt, but the matching requirement is loosening. AI augments the selection process, not the physical performance. |
| On-set safety coordination with team | 10% | 1 | 0.10 | NOT INVOLVED | Communicating with coordinator, other performers, camera operators, and SFX crew during stunt execution. Real-time physical coordination — hand signals, verbal cues, spatial awareness. Non-verbal trust signals. No AI involvement. |
| Administrative (call sheets, travel, scheduling) | 5% | 5 | 0.25 | DISPLACEMENT | Reviewing call sheets, managing personal schedules, travel logistics, paperwork. Structured tasks easily handled by AI scheduling and communication tools. |
| Total | 100% | 1.40 |
Task Resistance Score: 6.00 - 1.40 = 4.60/5.0
Displacement/Augmentation split: 5% displacement, 10% augmentation, 85% not involved.
Reinstatement check (Acemoglu): Minimal new task creation. CGI face replacement creates a minor new workflow where performers review and approve their digital likeness usage (per SAG-AFTRA 2023 contract). Virtual production LED wall stages require performers to adapt timing to digital environments. But the core work — physically executing dangerous stunts — is unchanged and will remain so. The role is stable, not transforming.
Evidence Score
| Dimension | Score (-2 to 2) | Evidence |
|---|---|---|
| Job Posting Trends | 0 | BLS projects "little or no change" for actors/performers (SOC 27-2011, 57,000 jobs, ~6,300 annual openings 2024-2034). Stunt performers are a tiny niche subset not tracked separately. Market is stable but extremely small — casting is relationship-driven, not posting-driven. |
| Company Actions | 0 | No studios cutting stunt departments citing AI. SAG-AFTRA's 2023 contract secured AI protections including consent and compensation for digital likenesses. Major productions continue employing full stunt teams. Digital doubles supplement but do not replace physical performers on set. |
| Wage Trends | 1 | SAG-AFTRA theatrical minimum $1,246/day (2025-2026). Mid-level performers earning $1,000-$3,000/week. Streaming platform competition for action content has pushed rates modestly upward. Union minimums provide a wage floor that tracks above inflation. |
| AI Tool Maturity | 1 | CGI face replacement reduces the appearance-matching premium for doubles, but no AI tool performs physical stunts. Disney's Stuntronics robots handle simple aerial maneuvers at theme parks but are nowhere near film-set complexity. Anthropic observed exposure for SOC 27-2011 (Actors) is 10.11% — very low. Core physical work has no viable AI alternative. |
| Expert Consensus | 0 | Mixed signals. Concerns exist about digital doubles and body scans enabling AI replication. But consensus is that physical stunt work persists — audiences prefer authentic action, and SAG-AFTRA protections guard against unconsented digital replacement. The role is evolving (CGI augmentation of face/appearance), not disappearing. |
| Total | 2 |
Barrier Assessment
Reframed question: What prevents AI execution even when programmatically possible?
| Barrier | Score (0-2) | Rationale |
|---|---|---|
| Regulatory/Licensing | 1 | SAG-AFTRA registration required for union film/TV work. Production insurance mandates human stunt performers on set. Some jurisdictions require safety certifications for pyrotechnics and rigging work. Not as strict as medical licensing but a meaningful professional framework. |
| Physical Presence | 2 | Essential and non-negotiable. The performer's body IS the product — falling from height, fighting, burning, crashing vehicles. Each environment is unstructured and unpredictable. All five robotics barriers apply at maximum: dexterity in varied terrain, safety certification for dangerous environments, liability, cost economics, cultural trust. No remote or automated alternative exists. |
| Union/Collective Bargaining | 2 | SAG-AFTRA represents stunt performers in film/TV. The 2023 contract explicitly protects against AI replacement and requires consent/compensation for digital likenesses. The 118-day strike demonstrated the industry's willingness to halt all production over AI protections. Union agreements define crew minimums, rates, and working conditions. |
| Liability/Accountability | 1 | The stunt coordinator bears primary safety liability, but production companies carry insurance requiring human performers. If a performer is injured, the production faces legal and insurance consequences. However, the performer themselves is not the primary liability-bearer — that falls to the coordinator and production. |
| Cultural/Ethical | 1 | Audience and director preference for practical stunts — the knowledge that real humans performed dangerous action adds perceived authenticity and visceral impact. "No CGI" is a marketing selling point for action films. SAG-AFTRA strike demonstrated cultural resistance to AI replacement. But audiences are increasingly comfortable with CGI-enhanced action in superhero and fantasy genres. |
| Total | 7/10 |
AI Growth Correlation Check
Confirmed at 0 (Neutral). AI adoption does not change demand for stunt performers. Demand is driven by audience appetite for action content, streaming platform investment, and production budgets — all independent of AI trends. CGI face replacement changes which performers get hired for doubling work (less emphasis on physical resemblance) but does not reduce the total number of performers needed to execute physical stunts.
JobZone Composite Score (AIJRI)
| Input | Value |
|---|---|
| Task Resistance Score | 4.60/5.0 |
| Evidence Modifier | 1.0 + (2 × 0.04) = 1.08 |
| Barrier Modifier | 1.0 + (7 × 0.02) = 1.14 |
| Growth Modifier | 1.0 + (0 × 0.05) = 1.00 |
Raw: 4.60 × 1.08 × 1.14 × 1.00 = 5.6635
JobZone Score: (5.6635 - 0.54) / 7.93 × 100 = 64.6/100
Zone: GREEN (Green >=48, Yellow 25-47, Red <25)
Sub-Label Determination
| Metric | Value |
|---|---|
| % of task time scoring 3+ | 15% (body doubling 10% + admin 5%) |
| AI Growth Correlation | 0 |
| Sub-label | Green (Stable) — AIJRI >=48 AND <20% of task time scores 3+ |
Assessor override: None — formula score accepted. The 64.6 places this role 16.6 points above the Green threshold, a comfortable margin. Sits just above Stunt Coordinator (62.8) — appropriate because the performer spends a higher proportion of time on irreducible physical work (85% not involved vs 45%), yielding higher task resistance (4.60 vs 4.25), offset by lower barriers (7 vs 8, since coordinator bears primary liability) and lower evidence (2 vs 3). Consistent with Circus Performer (60.0 GREEN Stable) and Dancer (56.7 GREEN Transforming) — physical performance roles with strong Moravec's Paradox protection.
Assessor Commentary
Score vs Reality Check
The 64.6 Green (Stable) label is honest. The score is driven by extreme task resistance (4.60) — 85% of task time involves physical work that no AI or robot can perform on a live film set. Barriers (7/10) reinforce the classification with strong union protection and mandatory physical presence. The role is not borderline — 16.6 points above the nearest zone boundary. Even if barriers dropped to 3/10 and evidence turned neutral (0/10), the score would remain above 48. The classification is task-resistance-driven, not barrier-dependent.
What the Numbers Don't Capture
- CGI face replacement eroding the doubling premium. Deepfake face replacement technology means performers no longer need to physically resemble the lead actor. This changes WHO gets hired (skill over appearance) but not HOW MANY performers are needed. The performer whose primary value was looking like Tom Cruise is more at risk than the performer whose value is executing a 100-foot high fall.
- Tiny addressable market with extreme income inequality. Perhaps 2,000-3,000 active stunt performers in the US. Work is project-based with gaps between productions. Green Zone does not mean stable income — it means the work that exists is protected from AI. Most mid-level performers earn $46K-$118K annually with significant periods between bookings.
- Physical career lifespan. Stunt work is physically punishing. Injuries accumulate. Most performers have a 15-20 year physical window. AI risk is not the primary career threat — the body's limits are.
- Geographic concentration. Work is concentrated in Los Angeles, Atlanta, Vancouver, and London. Outside production hubs, opportunities are virtually non-existent regardless of AI risk.
Who Should Worry (and Who Shouldn't)
Performers whose value is in what they DO are safer than this score suggests. If your specialisation is high falls, precision driving, fire work, or complex fight choreography — skills that require years of physical training and cannot be faked — you are in one of the most AI-proof positions in entertainment. The more dangerous and physically demanding the stunt, the more protected you are.
Performers whose primary value was physical resemblance to a lead actor should pay attention. CGI face replacement means productions can now hire the best athlete for the stunt and digitally apply the lead actor's face in post. This is good for skilled performers and bad for look-alikes. The single biggest separator: whether your value is in physical skill execution or in looking like a star. Skill is permanently protected; appearance matching is eroding now.
What This Means
The role in 2028: The mid-level stunt performer still falls from buildings, fights on camera, burns in protective gel, and crashes vehicles — exactly as they do today. CGI face replacement is standard on major productions, meaning the performer is hired for physical skill rather than actor resemblance. Virtual production LED wall stages change the visual backdrop but the physical stunt is identical. AI tools help the coordinator design and pre-visualise sequences, but the performer's job — executing dangerous physical action — is unchanged.
Survival strategy:
- Deepen physical specialisation. The more dangerous and technically demanding your stunt discipline, the more protected you are. High falls, precision driving, fire burns, and complex wire work are the hardest to replace and the most valuable to productions.
- Maintain SAG-AFTRA standing and safety record. Union protections are your structural moat. An impeccable safety record and strong relationships with coordinators are career currency that no AI can replicate.
- Embrace CGI integration. Learn to work with virtual production stages, understand how face replacement works, and be comfortable with 3D body scans (with appropriate SAG-AFTRA consent protections). Performers who bridge physical and digital workflows will be the most employable.
Timeline: 15-25+ years before any meaningful automation reaches physical stunt execution on film sets. Driven by Moravec's Paradox — performing a high fall in an unstructured environment is extraordinarily hard for robots — combined with union protections, production insurance requirements, and audience preference for authentic action.
