Graphic designers score 86% on Oxford Martin School's 2013 automation scale — placing the profession in the high-risk category. Anthropic's 2026 Economic Index shows 37% observed AI exposure, meaning over a third of professional AI usage in design contexts is already AI-mediated. Our combined score is 56% — firmly high risk. The theoretical ceiling is 86%. The observed floor is already 37%. Check your specific design role here →
This surprises people. Design feels creative, human, expressive. Surely it is one of the roles AI cannot touch. The research says otherwise — and if you work in visual design, you have almost certainly already felt why.
What the 86% is measuring
The Frey and Osborne methodology assessed automation risk based on the tasks that actually make up an occupation in practice. For graphic designers, a large proportion of daily work involves pattern application, template execution, format adaptation, and iterative visual production — tasks that follow rules and conventions rather than generating genuinely novel solutions.
The score was calculated in 2013. What nobody anticipated was how quickly AI image generation would arrive, and how directly it would target exactly those tasks.
What has already happened to the design market
AI image generation — Midjourney, DALL-E, Stable Diffusion and their successors — arrived at scale in 2022 and 2023. The impact on specific design markets was immediate and significant.
Stock photography agencies saw licensing revenue collapse as AI-generated images replaced what would previously have been purchased stock. Illustrators reported losing clients who had previously hired them for book covers, editorial images, and marketing materials. One illustrator described spending hours learning to draw in an AI style to meet a client’s reference image — an AI-generated image the client had used as a brief. A third of UK illustrators reported losing commissions to AI within two years of its mainstream arrival.
Anthropic’s March 2026 research confirmed the pattern: creative and visual production roles have among the highest observed AI exposure of any category in the data. The 86% is not a theoretical future — it is already visible in hiring patterns and freelance market rates.
Where AI is weakest in design
The parts of design that AI continues to struggle with are the parts that involve genuine conceptual thinking, strategic communication, and the kind of cultural intelligence that comes from understanding a client’s audience at a human level.
A brand identity is not a logo. It is the distillation of what a company is, who it speaks to, and how it wants to be perceived — and that requires understanding that goes beyond visual pattern recognition. UX design is not visual styling. It is the architecture of how humans interact with systems, which requires deep understanding of human behaviour, user research, and the ability to translate complex insights into clear design decisions.
The market collapse has happened at the commodity end of design — stock images, template-based production work, straightforward briefs with clear deliverables. The strategic and conceptual end is more resilient. The problem is that most designers spend most of their time at the commodity end, not the strategic one.
The honest picture for designers in 2026
The 86% score reflects something real about how design work is actually distributed in practice. It is a high-risk number, and it deserves to be treated as one.
For designers whose work is primarily production — adapting assets, creating variations, executing briefs with defined parameters — the risk is immediate and it is already materialising. Clients are replacing this work with AI tools, and the market for it is contracting.
For designers who have moved into genuinely strategic territory — creative direction, brand strategy, UX research, the kind of work that requires deep understanding of human behaviour and business context — the picture is harder to call. This work is more resilient, but the competition for it is increasing as designers displaced from production work move toward it.
The honest answer is that design is not being replaced as an entire discipline. But the discipline is being fundamentally restructured, and the 86% score reflects the reality that a large share of what graphic designers actually spend their time doing is already being automated at speed.
Check your specific role
Where does your design role actually sit?
Enter your occupation and see your AI automation risk score, based on Oxford Martin School + Anthropic Economic Index — 758 occupations.
Check my job risk →Related
My Sister Was a Translator for 15 Years. AI Took Her Work in Months. →The Most At-Risk Jobs Right Now →Creative Industries — AI Risk Overview →Take the 2-minute quiz to assess your own risk →Based on Oxford Martin School research (Frey & Osborne, 2013) and Anthropic Economic Index (March 2026).