Â鶹´«Ã½'s Kathy Pham and MIT researcher Dr. Danielle Li discuss a critical challenge in the AI-powered enterprise: as AI models scale the expertise of top performers, organizations must rethink the implications of their strategic contributions.
Emily Faracca
Multimedia Content Writer
Â鶹´«Ã½
?
Audio also available on and .
In an AI-powered world, the true gold lies not just in data, but in the hidden wisdom of your workforce.
Imagine the top salesperson, the most effective customer service agent, or the most brilliant programmer in your organization. Their work is invaluable, not just for the impact that they make, but for the hard-won, unarticulated expertise that fuels their success.
Now, imagine a machine is watching their every move.
The conversation around AI often revolves around task automation. But the true, deeper promise of AI is not in merely replacing tasks, but in its ability to serve as a shared repository¡ªa collective well¡ªfor human expertise.
This ability to scale the knowledge of your most seasoned employees and democratize it across the entire organization is creating a profound paradox at the heart of the modern organization.
It is challenging our foundational assumptions about productivity, compensation, and what it means to lead.
In a recent Future of Work podcast conversation, Kathy Pham, Â鶹´«Ã½¡¯s vice president of AI, sat down with Dr. Danielle Li, an AI researcher at MIT and member of Â鶹´«Ã½¡¯s AI Advisory Board, to discuss her latest research on this very phenomenon: How does the act of training an AI model impact the compensation and career path of a company's top performers?
Their discussion unveiled a blueprint for business leaders, highlighting the urgency of redefining work, redesigning compensation models, and embracing new emerging leadership mindsets.
Report
For years, organizations have struggled with the problem of "scattered expertise". The deep, contextual wisdom that makes a top performer excellent¡ªthe intuition, the judgment, and the unarticulated "how-to"¡ªis extremely difficult to share. When an organization grows, this knowledge is often fragmented, leading to uneven learning and performance across teams.
AI changes this equation by acting as an observer and a shared repository. By watching employees work and monitoring the resulting output, an AI model can extract the actions that make a person successful in a way that the person themselves might find difficult to explain.
The power of this system is that any new effective way of working is immediately captured by the model and shared with everyone in the organization, making the learning process collective and ongoing. It serves as a communal place for experience, moving the organization past the limitations of human memory and knowledge-sharing friction.
By watching employees work and monitoring the resulting output, an AI model can extract the actions that make a person successful in a way that the person themselves might find difficult to explain.
Li's in the customer service domain offers a perfect example of the AI paradox.
In her study, integrating an AI tool significantly increased overall productivity, with the greatest gains seen among the least experienced and lowest-skilled workers.
This acceleration meant that a newer employee with two months of tenure using the AI performed as well as an employee with six months of tenure working without the tool. The AI was effectively an "exoskeleton," disseminating the best practices of top performers and embedding mentorship directly into the workflow.
But here¡¯s the paradox:
The challenge is that the top performer's unique wisdom is extracted and then deployed to support the capabilities of others.
As Li noted, while skills used to live solely within our bodies, AI is capable of observing and extracting that expertise. The top performer is now inadvertently training a competitor.
As Li noted, while skills used to live solely within our bodies, AI is capable of observing and extracting that expertise. The top performer is now inadvertently training a competitor.
To overcome the reluctance of top performers to share their valuable data, companies must move away from a traditional, top-down approach and build a truly collaborative relationship. This requires a radical rethinking of compensation and how we define productivity.
In an AI-powered organization, productivity is no longer just the immediate act of doing the work (serving a customer or writing a report). It is a dual-layered effort: the act of working and the secondary act of providing examples and data to train and improve the AI model.
This secondary contribution¡ªthe "strategic work" of teaching the model¡ªmust be recognized and compensated.
To foster this co-creation mindset, companies should consider:
Transparency and Recognition: Being transparent about how employee data is used to build AI models and providing non-monetary recognition for that contribution.
A "Co-Creator" Mindset: Encouraging top performers to feel like a co-creator in the AI development process, which motivates them to proactively offer information to refine the system.
Credit as a Career Asset: Enabling employees to put their contribution to a specific model on their resume¡ªa form of professional ownership that they can carry with them, similar to a software engineer maintaining a GitHub repository.
As the boundaries of expertise and productivity blur, a new form of leadership is required to manage this fluid environment: the Chief Work Officer (CWO). This role is not simply an extension of the c-suite, but a new, cross-functional mindset dedicated to orchestrating how humans and AI collaborate across the entire organization.
The CWO is essential for seeing the total picture of "work,¡± whether it originates from a human body, a recording, or a trained AI model, as an integrative whole. This integrated view is critical for building flexible compensation and policy models that accurately reward all forms of contribution.
The key responsibilities for this emerging leader or cross-functional team include:
Decomposing Work: Thoughtfully analyzing and dividing work between humans and AI.
Workforce Readiness: Ensuring every employee is ready for success in the environment of human-machine collaboration.
Measuring the Unseen: Developing robust data infrastructure to measure and credit all components of work¡ªnot just the final output (e.g., a report or a slide deck), but the processes, time, and, critically, the training data provided to AI models.
Focusing on Strategic Value: Ensuring AI is implemented to unlock strategic growth and not merely for short-term cost reduction.
Ultimately, the CWO is not about managing AI itself, but about actively designing the future of work to be productive, ethical, and fulfilling.
The CWO is essential for seeing the total picture of "work,¡± whether it originates from a human body, a recording, or a trained AI model, as an integrative whole. This integrated view is critical for building flexible compensation and policy models that accurately reward all forms of contribution.
Beyond the strategic and financial implications, the integration of AI has a direct and profound impact on culture. AI can directly combat a major source of employee dissatisfaction: the unpleasantness of struggling to succeed at work.
In the case of the customer service study, the AI system improved the overall work experience for employees, leading to lower turnover. By offering suggestions to calm an irate customer, for instance, the AI improved people's social skills, leading to fewer requests for a manager and reducing the frequency with which an agent was faced with increasing customer frustration.
The AI served as a protective shield, supporting workers in the most difficult and undesirable parts of their jobs. By thinking of AI as a tool that is here to make our lives easier and our work more pleasant, leaders can foster a culture that embraces change and encourages a collaborative, rather than fearful, relationship with the technology.
In conclusion, the journey into the AI-augmented enterprise demands a new social contract between a company and its high-performing employees.
By adapting our measures of productivity and embracing new mindsets, organizations can navigate the AI paradox and design a future of work where human ingenuity is amplified.
Over half of business leaders are concerned about talent shortages¡ªand only 32% are confident their organization has the skills needed for success. See how AI is transforming skills management in .
Report
62% of leaders welcome AI, but only 55% of employees feel the same way. Closing this AI trust gap and ensuring that your organization is prepared for widespread AI implementation will prove decisive to your company's future success.