Research / Thesis brief
AI Competence 2035
A practical taxonomy linking AI issues to concrete competence choices.
AI preparedness · Organizations · Human judgment
01 / Why it matters
AI is moving fast.
Skills and judgment must keep up.
AI changes work, decisions, and public services faster than many institutions can adapt.
The research questionWhich competences should organizations build now to stay responsible and effective by 2035?
02 / Issues → competences
Start with the issue.
Find the response.
Competences are derived from issues: the Issue Space diagnoses challenge dimensions, and competence categories define organizational responses.
01DomainSector expertise+
Domain-Specific Aspects
Context fit of AI with sector workflows and knowledge.
Sector-Specific Domain Expertise
Industry and task understanding to make AI relevant in context.
02TechnologyTechnological competence+
Technical and Technological Aspects
Data quality, integration, infrastructure, and cybersecurity.
Technological (or Material) Competences
Technical and infrastructural foundations to build and run AI.
03LeadershipStrategic & organizational+
Managerial Leadership
Strategic alignment, culture, resources, trust, and leadership support.
Strategic and Organizational Competences
Ongoing alignment between AI initiatives, goals, and change capacity.
04IntelligenceCognitive competence+
Organizational Intelligence
Learning, adaptation, readiness assessment, and resource configuration.
Cognitive Competences
Organizational learning, sensemaking, and adaptive decision capability.
05RelationshipsInteractional competence+
Relationships and Networking
Human–AI interaction, cross-functional coordination, and stakeholder alignment.
Interactional Competences
Collaboration and coordination among humans, AI systems, and stakeholders.
06EthicsEthical & societal+
Ethical and Wider Impacts
Agency, labor effects, bias, regulation, and societal concerns.
Ethical and Societal Competences
Ability to anticipate and handle legal, ethical, and social impacts.
This mapping is a starting structure. In practice, competences overlap and must be bundled dynamically across contexts and transformation stages.
03 / A three-level challenge
One transformation.
Three scales.
As AI guidance expands, human discretion in professional judgment can narrow, creating tension between efficiency and responsible decision-making.
Organizations may become dependent on a few platforms for models, data, and tooling, which can reduce strategic autonomy and bargaining power.
The speed of AI deployment can outpace regulation, institutional learning, and social adaptation, amplifying systemic risk.
04 / How it was studied
Mixed methods,
one clear logic.
Map the issue space
Systematic review of AI transformation issues in management and policy contexts.
Refine the taxonomy
Semi-Delphi expert dialogue to test, adjust, and strengthen competence categories.
Check on real-world signals
International survey and complementary empirical analysis across institutional settings.
05 / Key findings
Three ideas
to take with you.
Competence is not only technical
Governance, discretion, ethics, and coordination are as critical as model-building skills.
Different sectors need different mixes
There is no one-size-fits-all AI skill model; context and institutional role matter.
Platforms shape capabilities
Platform choices influence what organizations can learn, govern, and scale over time.
About this brief
This page summarizes Shengxing Yang’s doctoral research at Université Paris-Saclay, defended on 20 February 2026. The full thesis is not currently available for public download.
The Issue Space offers six dimensions for examining AI transformation. The practitioner book and the PTSD reading companion extend this work into questions for practice; they are distinct from the doctoral thesis.