Shuichiro Ogawa
日本語

Notes · updated 2026-06-28

Designer Careers That Will Generate Value over the Next Five Years — Academic Review (2026)

An integrative summary of 37 peer-reviewed papers and major preprints collected through lightweight scoping. Complete bibliographic entries with DOI/URLs are in the “References” section below. Internal working ledger: source/review/designer-career-paths-ai/papers.md (not published). Industry side: designer-career-value-industry. Related: designer-role-ai-roundtable (industry x academia roundtable) / democratization-recommodification-paradox (paradox of democratization) / ai-expectation-gap-ux-literature (expectation gap) / ai-in-design-literature (AI x design academic review).

Survey Meta-Information

  • Collection date: 2026-06-28 / Count: 37 (lightweight scoping)
  • Venue emphasis: ACM CHI / DIS / CSCW / FAccT / C&C, Design Studies / She Ji / CoDesign, ISR / JVB / WES
  • Confidence notes: 4 preprints (P28, P34, P35, P37). Author names etc. [requires primary verification] for 3 items (P11, P14, P15). No peer-reviewed empirical research on DesignOps was found (addressed on the industry side).

TL;DR

The academic research identifies five structural changes.

  1. Role transition: Designers’ work shifts from “making” to “directing.” GenAI takes on the generation phase of divergence, and designers shift their center of gravity to curation, evaluation, and quality judgment (P14, P07).
  2. Double-edged effect on creativity: GenAI improves everyone’s creativity during ideation but impairs experts’ efficiency during implementation and increases design fixation (P06, P07).
  3. De-skilling risk for juniors: Cognitive offloading by AI inhibits the formation of foundational skills. Seniors view GenAI as supplementary, while juniors are directly concerned about unemployment risk (P03, P12).
  4. Redefinition of core competencies: Design judgment, coping with uncertainty, and quality evaluation through aesthetic knowledge emerge as core capabilities that are difficult for AI to replace (P15, P16).
  5. Freelance market displacement: A difference-in-differences analysis using ChatGPT’s launch as a shock confirmed clear displacement effects in the design freelance market (P34).

Key Findings by Domain

A. Role Transformation --- From Executor to Curator (7 papers)

Role transformation of designers is the finding where the most empirical studies converge (for the structural analysis, see maker-to-editor-paradigm).

el Kordy et al. (P14, CHI EA 2025) interviewed 17 strategic design professionals and showed that GenAI functions as a creative partner during the divergence phase but remains a tool during convergence and evaluation. The designer’s role transitions from “execution” to “direction and curation.” A FAccT 2026 study (P15) described practices in which designers exercise aesthetic knowledge to evaluate and integrate outputs and maintain trustworthiness during workplace AI transformation. Li et al. (P03, CHI 2024) conducted 20 interviews showing that experienced designers view GenAI as supplementary, while juniors directly fear skill degradation and unemployment. This asymmetry of perception reveals how AI’s significance varies by career stage. Takaffoli et al. (P04, DIS 2024) surveyed 24 practitioners and found that on-the-ground GenAI use is centered on document creation tasks, with limited application to design-specific tasks (wireframes, etc.). Clarke & Joffe (P35, 2025) described how creative agency workers actively reconfigure the division of labor with GenAI, introducing the concept of “interpretive template trust.” The simple replacement/augmentation dichotomy fails to capture this.

Career implications: The ability to judge “what to select from AI output, what to discard, and how to integrate it” becomes the differentiating factor in careers, rather than “what can be produced with AI.”

B. Double-Edged Effect on Creativity (4 papers)

GenAI’s impact on creativity is not unidirectional.

Hou et al. (P07, ISR 2025) combined controlled and field experiments to show that GenAI improves everyone’s creativity during ideation but impairs experts’ efficiency during implementation. Stage- and expertise-dependent usage is necessary. Wadinambiarachchi et al. (P06, CHI 2024) conducted a between-subjects experiment with 60 participants, reporting that groups using AI image generation exhibited increased design fixation, with decreases in the number, diversity, and originality of ideas. This is a direct counterargument to the optimistic claim that “GenAI expands creativity.” Thoring et al. (P08, ICED 2023) proposed 10 research questions for “the augmented designer” based on a typology of design knowledge. Shi et al. (P20, CSCW 2023) conducted an SLR of 93 papers and concluded that AI augments rather than replaces the creative process.

Career implications: Designers who use AI for idea divergence while maintaining the judgment to avoid fixation can preserve both breadth and depth of exploration. Stage-appropriate usage becomes a skill in itself.

C. De-Skilling Risk and the Junior Crisis (3 papers)

Among the paradoxical risks AI introduces, the impact on juniors is identified as most severe by multiple studies.

Shukla, Bui & Parsons (P12, CHI EA 2025) showed that while AI use increases task efficiency, paradoxical risks of de-skilling, cognitive offloading, and responsibility transfer become manifest (the literature map on cognitive offloading and learning). For those in the process of forming foundational skills, AI “shortcuts” inhibit skill acquisition itself. Li et al. (P37, 2025) investigated how vibe coding reshapes the design process through 22 interviews, describing tensions among de-skilling, ownership, and creativity protection. Ge & Fan (P18, E&PDE 2024) organized the role of AI in design education through an SLR of 35 papers, finding that while AI adoption in educational settings is surging, gaps in perception and practice exist on both the teacher and student sides.

Career implications: The greatest risk for juniors is not having their jobs taken by AI but losing the opportunity to develop foundational design judgment through using AI. How to design the balance between foundational acquisition and AI utilization is the challenge for education and early careers.

D. Redefinition of Core Competencies (5 papers)

Research empirically exploring which capabilities are difficult for AI to replace.

Shukla, Bui & Parsons (P16, C&C 2025) tracked 10 UX practitioners over 4 weeks, elucidating coping strategies for uncertainty (adaptive framing, negotiation, judgment). Design judgment itself is the core competency. Tan (P10, Design Studies 2021) synthesized research on design expertise, presenting four themes as foundations: experiential knowledge, adaptability, insight, and motivational support. Stige et al. (P01, IT&P 2024) concluded through an SLR that UX roles are shifting toward empathy and soft-skill dominance. Zdanowska & Taylor (P02, CHI 2022) demonstrated that non-technical UX practitioners exhibit high capability in ML system design. User-understanding capabilities remain effective in the ML era, not just technical knowledge. Gorichanaz (P13, PACMHCI 2025) identified 5 factors that impede HCD practice in organizations. Speed and direction clarity are the governing dimensions, suggesting the importance of designers’ organizational negotiation capabilities.

Career implications: Capabilities difficult for AI to replace converge on four areas: design judgment, coping with uncertainty, empathy, and organizational negotiation. These are practical knowledge formed through experience and reflection, not technical skills.

E. Transformation of the Creative Labor Market (4 papers)

Research capturing change at the market level.

Liu et al. (P34, arXiv 2023) used a difference-in-differences analysis with ChatGPT’s launch as a shock and demonstrated clear displacement effects in text- and design-related freelance markets. Those with AI-adaptive skills benefit, but those who cannot adapt are directly excluded. Erickson (P32, Creative Industries Journal 2024) presented 6 case studies showing that AI products are more labor-intensive than traditional media and that human contributions are invisibilized in the final product. Alacovska et al. (P33, WES 2024) surveyed 49 creative workers in the platform economy, identifying “relational work” as resistance to commodification, pre-precarity, and algorithmic norms. Bankins et al. (P36, JVB 2024) synthesized 104 papers in an SLR using career stage theory to integrate AI’s impact and presented a sustainable career perspective.

Career implications: In the freelance market, displacement by AI has been quantitatively confirmed. Relationship building, trust formation, and contextual understanding within organizations --- capabilities that are difficult to commodify on platforms --- become the defensive line.

F. Design Thinking Critique and Beyond (3 papers)

Lee (P26, She Ji 2021) critiqued design thinking within organizations for becoming fixated on a “production paradigm,” relying on individual agency to the point of impeding organizational transformation. Even if AI further streamlines this “production,” the organization’s problem structure does not change. Matthews et al. (P27, CoDesign 2023) described the challenge of co-design for wicked problems, demonstrating the difficulty of institutional transformation. van der Maden et al. (P28, arXiv 2025) identified five stop signs that stall GenAI discourse (trustworthiness, IP, tool framing, environment, economic degradation).

Career implications: Career paths that cultivate production skills under the banner of “design thinking” are most susceptible to AI replacement. Capabilities related to system-level problem framing, institutional transformation, and organizational decision-making are positioned above production.

Academic Sketch of the Designer Profile That Will Gain Value in the AI Era

The direction that academic literature indicates will “generate value over the next five years” can be organized into three tiers.

Tier 1 (most difficult to replace): Problem framing, strategic judgment, organizational transformation

  • The capability to decide what to build and what not to build
  • Structuring wicked problems, negotiating among stakeholders, institutional design
  • Design leadership, design management (P30: design expertise in top management mediates outcomes)

Tier 2 (value amplified through AI collaboration): Quality judgment, curation, human understanding

  • Evaluating, selecting, and integrating AI outputs (P14, P15)
  • User research, empathy, contextual understanding (P01, P02)
  • Design judgment under uncertainty (P16)
  • Designing human-AI interaction (P05, P25)

Tier 3 (large efficiency gains from AI, but difficult to differentiate on alone): Execution and production

  • Visual design, UI design, prototyping
  • Document creation, information architecture
  • Careers consisting only of this tier are experiencing ongoing displacement in freelance markets (P34)

Structural warning: The junior crisis cascades across all tiers. Tier 2 and Tier 1 capabilities are formed through Tier 3 experience, but if AI removes that experience, future seniors will not develop (P12). This is not merely an individual career problem but a question of reproducing the design profession as a whole. Related: the gaps in livelihood viability and initial entry (the entry-level gap).

Unresolved Issues

  1. Absence of longitudinal tracking: Nearly all studies are cross-sectional. No longitudinal research tracking career trajectories after AI adoption exists.
  2. Academic gap on DesignOps: No peer-reviewed empirical research on DesignOps was found. It is a growth area in industry but academic verification has not kept pace.
  3. Regional and cultural variation: The majority of collected literature comes from North America and Western Europe. Designer career transformation in Asia and emerging economies remains unexplored.
  4. Insufficient evidence for the “augmented designer”: P08 proposed research questions, but no empirical evidence of career outcomes from AI augmentation exists yet.
  5. Academic research on compensation and economic consequences: The only peer-reviewed study addressing AI’s impact on designer compensation or employment is P34’s freelance market analysis. Compensation changes for in-house designers remain academically unaddressed (for a review of the structural causes of compensation stagnation, see designer-salary-stagnation-literature).

References

← All Notes · Home