Managing Hybrid Intelligence: AI Integration and Human Decision-Making in Digital Workplace Transformation

Artificial intelligence is reshaping professional work, not by replacing human judgment, but by creating a collaboration between human and machine. This thesis examines how knowledge workers balance AI assistance with their own judgment, a phenomenon called hybrid intelligence, and identifies the factors shaping this balance. Drawing on survey data from 100 professionals across 18 industries and nine role categories, the study addresses six research questions on trust across task types, decision-making triggers, work outcomes, perceptions of human value, contextual influences, and role transformation since AI adoption. Findings challenge established assumptions. Trust in AI is moderate and stable across routine (M=3.32), analytical (M=3.31), and creative tasks (M=3.35), contradicting predictions of a task-based trust hierarchy. A paradox emerges: 94% frequently verify AI outputs, yet trust and verification are uncorrelated (r=–0.07), showing oversight is a professional norm. Trust decisions are expertise-driven: professionals trust AI most when they can critically evaluate outputs. Overrides follow complementary logic, triggered by task complexity, sensitivity, and generic outputs. Hybrid intelligence outcomes are positive: 85% report quality improvements, 77% efficiency gains, and 73% high confidence in outputs. Professionals consistently highlight emotional intelligence, creative originality, and ethical judgment as contributions AI cannot replicate. AI adoption elevates work: 56% report more strategic or creative roles, only 3% report increased routine work. These findings advance literature on trust in automation, hybrid intelligence design, and the augmentation-automation paradox, offering evidence-based guidance for organizations, trainers, and AI designers to make human-AI collaboration genuinely effective.