O*NET-SOC 41-9012.00

AI-Generated Models Are Here. Is Your Career Ready?

78 /100
Very high
Get my personal score, $9This score is for the occupation as a whole. The AI Risk Assessment moves it up to 15 points either way for how you spend your day, and names AI tools for your own tasks.

Models score 78 out of 100 (Very high) on the AIRRBridge AI exposure scale, built from the U.S. Bureau of Labor Statistics' Employment Projections 2025–35 and O*NET task data. Exposure measures how much of the work AI tools could change. It is not a forecast of job loss.

Generative AI tools can now produce photorealistic model imagery in seconds, and major brands, including Levi's and H&M have already run public campaigns using AI-generated likenesses in place of or alongside human models. That shift is accelerating, and it is hitting certain segments of the modeling market faster and harder than others.

Get your results by email

We'll send you a copy of this score, plus a heads-up if new BLS or O*NET data changes it.

Score updates only when the data changes. Unsubscribe in one click from any email.

How this score is built

BLS AI exposure: HighThe U.S. Bureau of Labor Statistics rates this occupation's relative exposure to AI as high. Exposure measures how much of the work AI tools could change, and BLS notes it does not imply job loss.62.5
O*NET task mixHow this occupation's tasks compare with other high-exposure occupations: routine, information-heavy work raises it, while caring, hands-on and judgment-heavy work lowers it.+11.1
BLS finding on AIIn its 2025–35 projections, BLS says: “Capital/labor substitution - share decreases as the use of artificial intelligence (AI)-generated models grows, reducing the need for models.”+5.0
ScoreKept between 5 and 95. Low is under 25, Moderate 25 to 49, High 50 to 74, Very high 75 and up. The headline number drops the decimal (78.6 shows as 78), so a score never shows in a higher band than it earned.78.6

Which tasks AI touches most, and least

Researchers rated all 14 O*NET tasks for this occupation on whether AI language tools could cut the time they take by at least half. These are the tasks rated most and least exposed. The ratings sit beside the score above and don't change it.

Most exposed

  1. Report job completions to agencies and obtain information about future appointments. Directly exposed
  2. Record rates of pay and durations of jobs on vouchers. Exposed with added software
  3. Promote products and services in television commercials, on film, or in videos. Exposed with added software
  4. Apply makeup to face and style hair to enhance appearance, considering such factors as color, camera techniques, and facial features. Exposed with added software
  5. Pose as directed, or strike suitable interpretive poses for promoting and selling merchandise or fashions during appearances, filming, or photo sessions. Exposed with added software

Least exposed

  1. Follow strict routines of diet, sleep, and exercise to maintain appearance. Not exposed
  2. Dress in sample or completed garments, and select accessories. Not exposed
  3. Assemble and maintain portfolios, print composite cards, and travel to go-sees to obtain jobs. Not exposed
  4. Work closely with photographers, fashion coordinators, directors, producers, stylists, make-up artists, other models, and clients to produce the desired looks, and to finish photo shoots on schedule. Not exposed
  5. Pose for artists and photographers, with or without clothes. Exposed with added software

Task ratings: Eloundou, Manning, Mishkin and Rock, “GPTs are GPTs: An Early Look at the Labor Market Impact Potential of Large Language Models” (2023), rated in 2023 and shown with the human raters' label. Task statements from the O*NET Database. How we use them.

Job outlook

Median annual pay$48,470
Projected employment change, 2025 to 2035−2.1%
Openings per year, on averageabout 700
Typical education to enterNo formal educational credential

Source: U.S. Bureau of Labor Statistics, Employment Projections 2025–35.

What the score means

A 42% automation susceptibility score means modeling is in genuine medium-risk territory, not in immediate freefall, but facing enough structural pressure that the trajectory matters more than the current moment. The lower repetitive task score of 35% reflects that no two shoots, clients, or briefs are identical, which provides some protection, but the creative thinking score of 44% and decision complexity score of 43% both land in the mid-range, meaning AI tools can credibly approximate enough of the creative and judgment-based work to compete for lower-tier commercial bookings. The 71% social interaction score is the most meaningful buffer here, pointing to the client-facing, relationship-driven, and live-performance aspects of the role that remain resistant to automation.

What AI is changing

AI model image generation replacing commercial shoots: Tools like Midjourney, Adobe Firefly, and DALL-E 3 are now capable of producing photorealistic human figures in branded contexts, and companies including Levi's partnered with Lalaland.ai in 2023 to generate diverse model imagery at scale without booking human models for every variation.

Virtual influencer and digital avatar campaigns: Brands like Prada, Calvin Klein, and Balmain have worked with fully synthetic digital models such as Shudu and Lil Miquela, demonstrating that AI-generated personas can carry commercial campaigns and accumulate social followings that rival human talent.

AI-powered fit and size modeling: Platforms like Reactive Reality and Veesual allow e-commerce retailers to generate on-model imagery across diverse body types and skin tones without casting real models, directly substituting for the large volume of catalog and e-commerce modeling work that makes up most mid-tier income in the profession.

Automated post-production reducing on-set requirements: AI retouching and body-editing tools such as Luminar AI and Adobe's generative fill have compressed the need for extensive on-set perfection, shifting more of the "model's job" into post-production and reducing the premium placed on physical precision during shoots.

AI casting and look-matching tools: Agencies and brands are beginning to use AI tools to identify models from databases based on visual criteria, reducing the relational and scouting work that once required human judgment, and making casting faster and more algorithmic at the commercial tier.

What stays human

Live performance and presence that cannot be faked: Runway work, live events, trade shows, and brand activations require a human being physically in the space, moving, reacting, and representing the brand in real time, and no AI-generated image can substitute for that kind of embodied presence, which keeps a significant segment of the profession structurally protected.

Client and creative team relationships: The 71% social interaction score reflects that modeling is not just about appearance but about collaboration, with photographers, art directors, stylists, and brand teams who develop preferences for specific talent they trust to deliver on a brief under pressure. Those working relationships are built over time and are not easily transferred to a synthetic asset.

Brand authenticity and consumer trust: A growing segment of consumers actively values knowing that a brand used real people, and the backlash to AI model campaigns has prompted several brands to publicly recommit to human talent. That consumer pressure creates a market floor for authentic human representation that AI cannot occupy.

High-end editorial and luxury fashion: At the top of the market, creative thinking at 44% still means that elite models bring a collaborative creative contribution, a signature quality of movement or expression, and a personal brand that luxury houses are paying for specifically. AI-generated imagery lacks the cultural cachet and editorial history that make top-tier talent commercially valuable.

Licensing, consent, and legal complexity: The legal landscape around AI-generated likenesses is still actively being written, and several jurisdictions are moving to require explicit consent for AI training on a person's image. That regulatory friction slows wholesale replacement and keeps human models relevant in any context where brand liability matters.

Where your skills transfer

Personal brand development and visual identity: Models who have built a recognizable personal brand, cultivated an audience, and managed their own public image carry skills that translate directly into roles like Brand Ambassador and Social Media Content Creator, both of which reward the ability to present consistently and compellingly across platforms.

Client communication and creative collaboration: The ability to interpret a brief quickly, take direction gracefully, and deliver under pressure in a room full of stakeholders is exactly what Commercial Actor and Art Director's Talent Liaison roles require, and models who have worked across multiple clients and campaign types develop this fluency naturally.

Posture, physical presentation, and training: The discipline of physical presentation, movement, and body awareness transfers well into Fitness Instructor and Corporate Presentation Coach roles, where teaching others to carry themselves with confidence and clarity is the entire product.

Industry network and casting relationships: Years of working relationships with agencies, photographers, stylists, and brand professionals provide a foundation for transitioning into Talent Agent or Casting Coordinator roles, where that network becomes the core professional asset rather than a byproduct of the work.

Adaptability and performance under variable conditions: The ability to show up ready, adjust to changing creative direction on the spot, and deliver consistent quality across unpredictable environments is highly valued in Brand Experience Specialist and Event Host roles, where no two engagements are the same.

Thinking about a move?

These occupations in the same field score at least 10 points lower for AI exposure, draw on the most similar core skills in O*NET's ratings, and need at most one step more schooling to enter. They're a starting point, not a record of where Models have gone.

Gambling Change Persons and Booth Cashiers44Moderate · median pay $36,220
Cashiers62High · median pay $32,880
Parts Salespersons66High · median pay $38,630

The Career Pivot Report goes further for $19, one time. It builds three paths from the skills you already use as one of today's Models: one AI-protected, one AI-enhanced and one AI-embraced. Each shows what carries over, what to add, recommended courses, projected salaries with cited sources and a 30-day start.

See the Career Pivot Report

Common questions

How will AI affect Models?

AIRRBridge rates Models at 78 out of 100 (Very high) for AI exposure. Exposure measures how much of the work AI tools could change, not whether the job will disappear. BLS projects employment in this occupation to decline 2.1% from 2025 to 2035, with about 700 openings a year. BLS also names AI as a reason demand for this occupation is expected to slow.

How much do Models make?

BLS reports a median annual pay of $48,470 for this occupation.

What education do Models need?

BLS lists no formal educational credential as the typical education needed to enter this occupation.

This page scores Models as a group. Your own job may score higher or lower. The AI Risk Assessment gives you your personal score, names the AI tools that fit your tasks and lays out a 90-day plan. $9, one time.

Get my personal score, $9

Not ready to buy? Build your resume free, with no account and no card. The industry is already splitting, so find out exactly where you stand. Check another occupation free.

O*NET-SOC 41-9012.00 · O*NET 30.2 data, scored October 2026. Source: U.S. Bureau of Labor Statistics, Employment Projections 2025–35. Includes information from the O*NET Database by USDOL/ETA, used under CC BY 4.0, modified by AIRRBridge. Scores are estimates for information only, not career advice.

Related occupations