This study examines the effects of Artificial Intelligence (AI) on knowledge worker productivity and quality in a realistic consulting context. The experiment involved 758 consultants from a global management consulting firm, who were randomly assigned to different conditions of AI access. The results show that consultants using AI were more productive and produced higher quality results in tasks within the capabilities of AI. However, for tasks outside the AI capabilities, consultants using AI were less likely to produce correct solutions. The analysis also reveals two distinct patterns of successful AI use by humans - some acting as “Centaurs” and others as “Cyborgs” - indicating different approaches to human-AI integration. Overall, the study highlights the potential benefits of AI augmentation in consulting tasks, while also recognizing the limitations and challenges in integrating AI into complex knowledge-intensive tasks.
| Signal | Change | 10y horizon | Driving force |
|---|---|---|---|
| Effects of AI on knowledge worker productivity | Increase in productivity and quality | Increased reliance on AI in tasks | Desire for efficiency and improved results |
| Capability of AI creates a “jagged technological frontier” | Uneven distribution of AI capabilities | AI will become more capable in some tasks | Advancement in AI technology |
| Consultants using AI were more productive and produced higher quality results | Improvement in task completion and quality | Increased use of AI in consulting tasks | Desire for improved performance and efficiency |
| Consultants using AI were less likely to produce correct solutions for tasks outside the AI frontier | Limitations of AI in certain tasks | Increased development of AI capabilities | Need for further advancements in AI |
| Emergence of two patterns of successful AI use: “Centaurs” and “Cyborgs” | Different approaches to human-AI integration | Integration of AI into task workflows | Exploration of optimal AI-human collaboration |