Role Definition
| Field | Value |
|---|---|
| Job Title | Door Supervisor / Bouncer |
| Seniority Level | Mid-Level (2-5 years, SIA licensed) |
| Primary Function | Controls entry to licensed premises (pubs, clubs, events). Checks IDs and verifies age, manages queues and crowd flow at entry points, de-escalates verbal confrontations, physically intervenes when necessary (ejections, restraining aggressive individuals), liaises with police and emergency services, provides first aid response, and maintains a visible deterrent presence. Works in unstructured, unpredictable, high-alcohol environments where every shift is different. |
| What This Role Is NOT | NOT a generic security guard (premises patrol/CCTV monitoring — scored separately at 43.6 Yellow). NOT a close protection officer (VIP bodyguard — personal protection detail). NOT an event steward (crowd management without door authority or SIA licence). NOT a loss prevention specialist (retail theft prevention). |
| Typical Experience | 2-5 years. Mandatory SIA Door Supervisor licence (Level 2 Award for Working as a Door Supervisor + Emergency First Aid at Work, renewed every 3 years with refresher training since April 2025). BLS maps to SOC 33-9032 Security Guards. UK: ~354,580 active Door Supervisor licences (71.3% of all 460,138 SIA licence holders, 2025). |
Seniority note: Entry-level (0-1 years, newly SIA-licensed) would score similarly — the physical presence and conflict management requirements exist from the first shift. Head Doormen/Security Supervisors (5+ years, managing door teams, venue security planning, client relationships) would score slightly higher — leadership adds judgment and coordination that further distances the role from automation.
Protective Principles + AI Growth Correlation
| Principle | Score (0-3) | Rationale |
|---|---|---|
| Embodied Physicality | 3 | The role is defined by physical presence at a venue entrance in chaotic, alcohol-fuelled, unpredictable environments. Ejecting aggressive patrons, breaking up fights, physically restraining individuals, managing crowd surges at 2am — every shift is different. Peak Moravec's Paradox. No robot is ejecting a violent drunk from a nightclub. |
| Deep Interpersonal Connection | 2 | Significant interpersonal component: reading body language and intoxication levels, de-escalating aggressive individuals through verbal authority and empathy, managing upset patrons denied entry, building rapport with regulars. Not therapeutic, but the human ability to defuse a confrontation before it becomes physical IS the core skill. |
| Goal-Setting & Moral Judgment | 1 | Real-time judgment on who to admit and refuse, when to intervene physically vs. verbally, when to call police vs. handle internally. Operates within venue policies and licensing law, but exercises significant discretion in ambiguous situations (suspicious behaviour, borderline intoxication, group dynamics). Less autonomous than police (no arrest powers, more bounded by venue rules). |
| Protective Total | 6/9 | |
| AI Growth Correlation | 0 | Neutral. AI adoption neither creates nor destroys demand for door supervisors. Demand is driven by the night-time economy (number of licensed premises, events, local authority licensing conditions) — not technology deployment. AI CCTV and ID scanners augment but do not change headcount requirements. |
Quick screen result: Protective 6/9 with neutral growth = Green Zone signal. Proceed to confirm with task decomposition and evidence.
Task Decomposition (Agentic AI Scoring)
| Task | Time % | Score (1-5) | Weighted | Aug/Disp | Rationale |
|---|---|---|---|---|---|
| Queue management & entry control | 25% | 1 | 0.25 | NOT INVOLVED | Managing the queue, controlling crowd flow, maintaining visible deterrent presence at the door, deciding entry pace based on venue capacity. Entirely physical, entirely human. You cannot automate the act of standing at a door, reading a crowd, and physically controlling who enters. |
| Conflict de-escalation & verbal management | 20% | 1 | 0.20 | NOT INVOLVED | Talking down aggressive individuals, managing group dynamics, calming disputes between patrons, refusing entry to intoxicated persons while preventing escalation. Reading micro-expressions, body language, and alcohol-altered emotional states in real time. The human ability to defuse a confrontation IS the value. Irreducible. |
| ID verification & age checking | 15% | 2 | 0.30 | AUGMENTATION | Checking IDs, verifying age, spotting fakes, matching face to photo. AI-powered ID scanners (e.g., IDScan, Scannet) can verify document authenticity and check banned lists. But the door supervisor still makes the final call — assessing whether the person matches the ID, reading behavioural cues of underage patrons using someone else's ID, and managing the human interaction. AI assists the check; the human decides entry. |
| Physical intervention & ejection | 15% | 1 | 0.15 | NOT INVOLVED | Physically removing aggressive, violent, or non-compliant individuals from the premises. Restraining combatants. Using proportionate force in chaotic, confined environments (stairwells, dance floors, smoking areas). No AI or robotic system can perform this. The SIA physical intervention training exists because this is hands-on, dangerous, human work. |
| Surveillance & venue monitoring | 10% | 3 | 0.30 | AUGMENTATION | Observing the venue interior for signs of trouble — drug use, fights developing, vulnerable individuals, overcrowding in specific areas. AI CCTV analytics can flag anomalies and alert staff, but the door supervisor still interprets context and responds physically. AI makes monitoring more efficient but doesn't replace the human who acts on the information. |
| Incident reporting & documentation | 5% | 4 | 0.20 | DISPLACEMENT | Writing incident reports, logging refusals, documenting use of force, maintaining door registers. Template-based documentation that AI can generate from dictation or body-worn camera footage. Small proportion of time, highly automatable. |
| Emergency & first aid response | 5% | 1 | 0.05 | NOT INVOLVED | First aid for injuries, managing medical emergencies (overdoses, collapses, glass injuries), fire evacuation, coordinating with ambulance services. Physical, hands-on, time-critical. No AI involvement. |
| Police & emergency services liaison | 5% | 2 | 0.10 | AUGMENTATION | Communicating with police about incidents, providing witness statements, coordinating during venue raids or emergency situations. AI communication platforms may streamline dispatch but human-to-human coordination with law enforcement remains essential. |
| Total | 100% | 1.55 |
Task Resistance Score: 6.00 - 1.55 = 4.45/5.0
Displacement/Augmentation split: 5% displacement, 30% augmentation, 65% not involved.
Reinstatement check (Acemoglu): Minimal new tasks. Some door supervisors now operate ID scanning systems and review AI-flagged alerts from CCTV analytics, but these are minor extensions of existing work, not new labour demand. The role is not transforming — it is persisting largely unchanged because the core function (human physical presence at a door) is irreducible.
Evidence Score
| Dimension | Score (-2 to 2) | Evidence |
|---|---|---|
| Job Posting Trends | 1 | UK SIA licence holders grew 6-7% from 2023 to 2025 (431,701 to 460,138). Door Supervisor licences now 71.3% of all SIA licences (up from 68.5% in 2023, 66.5% in 2022) — the fastest-growing licence category. Listed among top 5 in-demand SIA jobs for 2026. US BLS projects "little or no change" for Security Guards (33-9032) 2024-2034 with 162,300 annual openings, but door supervisor demand in the UK night-time economy is stronger than aggregate guard data suggests. |
| Company Actions | 0 | No security companies or venue operators are cutting door supervisor roles citing AI. ID scanning technology is being deployed across UK venues (20+ late-night venues in Oxford alone), but as a tool for door supervisors to use, not a replacement. The SIA refresher training mandate (April 2025) signals continued investment in the human workforce. No AI-driven headcount reductions observed. |
| Wage Trends | 0 | UK average: GBP 14.15/hr (Indeed, Feb 2026), GBP 26,441/yr (Glassdoor). London premium: GBP 30,861/yr. Experienced head doormen: GBP 32-34/hr. Wages are stable but not surging — tracking inflation rather than outpacing it. No AI skills premium; no downward wage pressure from automation. Low barriers to entry keep wages moderate despite physical risk. |
| AI Tool Maturity | 1 | AI ID scanners (IDScan, Scannet) verify document authenticity and cross-reference banned lists. AI CCTV analytics flag anomalies in venue interiors. Facial recognition being piloted in UK stadiums and some large venues (47% of stadiums list biometrics as 2025 priority). But all tools augment the door supervisor — none performs the core entry control, de-escalation, or physical intervention functions. No viable AI alternative exists for 65% of task time. |
| Expert Consensus | 1 | Universal industry consensus: AI augments, does not replace. SIA training updates focus on modern technology proficiency, not workforce reduction. Security industry publications (SDM Magazine, Security Base Group, Professional Security Magazine) consistently describe hybrid human-AI models. willrobotstakemyjob.com rates Security Guards at 42% calculated risk, but this aggregate figure masks the much lower risk for physical-confrontation roles vs. screen-monitoring roles. |
| Total | 3 |
Barrier Assessment
Reframed question: What prevents AI execution even when programmatically possible?
| Barrier | Score (0-2) | Rationale |
|---|---|---|
| Regulatory/Licensing | 2 | Mandatory SIA Door Supervisor licence required by law (Private Security Industry Act 2001). Level 2 qualification covering conflict management, physical intervention, and legal duties. Criminal record check, biometric registration, GBP 184 fee, 3-year renewal with mandatory refresher training (since April 2025). It is illegal to work as a door supervisor without this licence. No legal framework exists for licensing an AI system to control entry to licensed premises. This is a strong, statute-backed barrier. |
| Physical Presence | 2 | The strongest barrier. The entire role is defined by having a human body at the door of a venue. Ejecting violent patrons, physically restraining combatants, managing crowd surges, searching individuals — this is not an incidental feature of the role, it IS the role. No robotic system can operate in the chaotic, confined, alcohol-fuelled environments where door supervisors work (narrow stairwells, packed dance floors, wet pavements at 3am). |
| Union/Collective Bargaining | 0 | Door supervisors are overwhelmingly self-employed contractors or employed by security firms with no union representation. High turnover, fragmented employment, at-will engagement. No collective bargaining protections. |
| Liability/Accountability | 1 | Use-of-force decisions carry personal legal liability. Door supervisors can face criminal charges for excessive force, civil claims for injuries, and SIA licence revocation for misconduct. Venues carry duty of care obligations under licensing law. When a patron is injured during ejection, a human must be accountable. But liability is less acute than for police (narrower scope of authority, fewer lethal force scenarios). |
| Cultural/Ethical | 1 | Strong cultural expectation of human presence at venue doors. Patrons interact with door staff — negotiate entry, ask questions, seek help when feeling unsafe. The human doorperson provides reassurance and social authority that a camera or scanner cannot. Facial recognition at venue entrances faces significant privacy pushback (ICO guidance, GDPR, Big Brother Watch campaigns). Moderate cultural barrier. |
| Total | 6/10 |
AI Growth Correlation Check
Confirmed 0 (Neutral). AI adoption does not create more door supervisor demand and does not destroy it. The number of door supervisors needed is driven by the number of licensed premises, local authority licensing conditions (which often mandate minimum door staff numbers), event calendars, and the size of the night-time economy — not technology deployment. AI CCTV and ID scanners make existing door supervisors slightly more effective but create no new positions and eliminate no existing ones. This is Green (Stable), not Green (Accelerated) or Green (Transforming).
JobZone Composite Score (AIJRI)
| Input | Value |
|---|---|
| Task Resistance Score | 4.45/5.0 |
| Evidence Modifier | 1.0 + (3 x 0.04) = 1.12 |
| Barrier Modifier | 1.0 + (6 x 0.02) = 1.12 |
| Growth Modifier | 1.0 + (0 x 0.05) = 1.00 |
Raw: 4.45 x 1.12 x 1.12 x 1.00 = 5.5821
JobZone Score: (5.5821 - 0.54) / 7.93 x 100 = 63.6/100
Zone: GREEN (Green >=48, Yellow 25-47, Red <25)
Sub-Label Determination
| Metric | Value |
|---|---|
| % of task time scoring 3+ | 15% |
| AI Growth Correlation | 0 |
| Sub-label | Green (Stable) — <20% task time scores 3+, not Accelerated |
Assessor override: None — formula score accepted. The 63.6 sits comfortably in the Green zone, 15.6 points above the threshold. This correctly reflects a role where 65% of task time is completely untouched by AI and the remaining augmented/displaced time is peripheral to the core function. The score calibrates well against comparators: higher than Security Guard (43.6) because the door supervisor role is more confrontational and less screen-based, higher than Correctional Officer (49.5) because evidence is positive rather than negative, and comparable to Police Patrol Officer (65.3) which has stronger barriers but similar task profiles.
Assessor Commentary
Score vs Reality Check
The 63.6 Green (Stable) label is honest and well-calibrated. The role is not barrier-dependent — removing all barriers (setting to 0/10) would still produce a score of 56.7 (Green), because the task resistance alone (4.45/5.0) is exceptionally strong. Only 5% of task time faces displacement (incident reporting), and 65% is completely beyond AI's reach. The positive evidence (+3) reflects genuine demand growth in the UK night-time economy, not a temporary shortage. This is one of the most straightforwardly AI-resistant roles in the economy: the job is to physically stand at a door and decide who gets in, using your body, your voice, and your judgment. No technology changes that.
What the Numbers Don't Capture
- Local authority licensing mandates create a regulatory floor. Many UK licensing authorities require minimum numbers of SIA-licensed door staff as a condition of premises licences. A nightclub with a 500-person capacity might be required to have 4 door supervisors regardless of what technology is deployed. This structural demand guarantee is not captured in the evidence score.
- The night-time economy is the demand driver, not security technology. If pubs and clubs close (as during COVID), door supervisors lose work — not because of AI, but because there are no doors to supervise. The biggest risk to this role is economic, not technological.
- Stratification by venue type matters. Corporate event door staff at a gala dinner face different risks than nightclub bouncers in a city centre at 2am. The former is more ceremonial (lower physicality, more augmentable); the latter is more confrontational (higher physicality, less augmentable). This assessment scores the typical pub/club/event door supervisor.
- Self-employment and gig economy structure. Many door supervisors work as self-employed contractors through agencies, which means no employment protections but also high labour market flexibility — easy to enter, easy to leave. High turnover (first-time SIA licences declining from 94,332 in 2024 to 79,458 in 2025) may reflect attractiveness of the role more than demand.
Who Should Worry (and Who Shouldn't)
Door supervisors working the front door of busy pubs, clubs, and events — checking IDs face-to-face, managing queues, de-escalating confrontations, and physically intervening when needed — are among the safest workers in the economy. Your job is your body, your voice, and your judgment in an unpredictable environment. No AI system can do any of that. Door supervisors whose role has shifted toward static CCTV monitoring inside a venue or primarily administrative duties (logging, reporting) are slightly more exposed, but even they benefit from the SIA licensing barrier and the physical presence expectation. The single biggest separator: whether you are at the door dealing with people, or behind a screen watching footage. The door is safe. The screen is less so — but even screen-based door work involves far more human interaction than generic security guarding.
What This Means
The role in 2028: The door supervisor of 2028 uses an AI-powered ID scanner that flags fake documents and cross-references a banned patron database. Venue CCTV runs AI anomaly detection that alerts the door team to developing situations inside. But the door supervisor still stands at the entrance, reads the crowd, makes the call on who gets in, de-escalates the angry group denied entry, and physically removes the violent patron at closing time. The tools get smarter. The job stays the same.
Survival strategy:
- Maintain your SIA licence and complete refresher training promptly — the licensing barrier is your strongest structural protection, and lapsed licences mean lost work
- Develop advanced de-escalation and conflict management skills — these deeply human capabilities become the highest-value differentiator as venues invest in AI for the routine checks
- Learn to operate ID scanning and CCTV analytics systems — door supervisors who can use technology effectively become more productive and more valuable to employers, covering the augmentation trend rather than being replaced by it
Timeline: 15-25+ years before any meaningful displacement, if ever. The irreducible requirement for a licensed human body at the door of a licensed premises, combined with the chaotic, unpredictable, alcohol-fuelled environments in which door supervisors operate, makes this one of the most durable roles in the labour market.
