Artificial Intelligence
AI Needs You (More Than You Think)
MIT's Dr. Danielle Li discusses AI, skill augmentation, and the future of work.
Julie Colwell
Principal Strategist
Â鶹´«Ã½
The fear is familiar: AI edging humans out of the workplace. But Danielle Li, economist and MIT Sloan professor, sees something different in the data. Her research on hiring, innovation, and AI suggests that the real power of new technology emerges when it complements human expertise¡ªnot when it tries to replace it.
Li reports that when algorithms and employees work together, organizations get more from the knowledge, experience, and judgment of their people.
Li¡¯s landmark study, , doesn¡¯t just spark theory¡ªit¡¯s got practical implications for companies wrestling with their AI strategy. From decoding bias in NIH evaluations to how to leverage generative AI in customer service, Li explores how emerging technologies reshape productivity, hiring, and the future of jobs.
Report
C-Suite Global AI Indicator Report: AI Is the Ultimate Level-Up
AI and the New Talent Equation
For decades, hiring managers have searched for the perfect candidate¡ªsomeone with the right mix of technical knowledge, social acumen, and interpersonal skills. According to Li, AI is fundamentally changing that equation.
¡°In the past, you needed a combination of all three for each role, but AI relaxes these traditional constraints,¡± she explains. It provides support in areas where an otherwise promising employee might be lacking. A brilliant strategist with developing technical skills, can use AI tools to augment his or her expertise. Or a technical genius can use AI to learn to communicate better. This opens up a wider pool of talent.
This shift has a direct impact on career paths. As skills become more easily augmented, traditional career ladders begin to dissolve. Instead, organizations will see a more fluid workplace where employees can pivot between roles, acquiring new skills as they go.
For business leaders, this presents both a massive opportunity for internal mobility and a new challenge. As Li notes, leaders must now figure out how to manage an organization that has become a lot more dynamic and a lot more customized, which includes identifying people for new opportunities, putting them into those roles, and ensuring the process feels fair.
?As skills become more easily augmented, traditional career ladders begin to dissolve.
Li's research shows that AI doesn¡¯t just complete tasks, but also actively teaches. Her study on using an AI in customer service found that employees who used an AI assistant learned durable skills that they retained even when the tool was unavailable. The AI acts as a persistent mentor, offering a level of guidance that is practically infeasible for a human to replicate. Unlike a manager who provides high-level feedback once a week¡ªthe equivalent of telling a junior colleague to simply "work harder and connect more"¡ªthe customer service AI gave concrete examples of what improvement actually looks like in real-time, every day.
It nudged employees to try new approaches, like using more empathetic language, or responding to questions in ways that may be counter-intuitive to them, and in doing so, they learn what works.
Digital Replicas and Expert Collaboration
Looking ahead, Li is cautiously interested in the concept of a digital replica, a personalized AI model trained on the specific expertise of a top performer. Imagine capturing the unique surgical techniques of a world-class doctor or the strategic intuition of a seasoned executive and embedding that knowledge into an AI that can scale across an organization - or even a household.
¡°One of my colleagues played around with making a model of herself for her nanny to access. She trained it on her kids¡¯ allergies, preferences, schedules¡ everything she could think of that she knew would be useful,¡± Li explains. ¡°There¡¯s so much potential there.¡±
However, unlocking this potential hinges on a critical factor: collaboration. Organizations can¡¯t build an effective model by simply scraping company emails and public media. While a company could create a basic model without an employee's consent, Li cautions that it won¡¯t be as valuable without the top performers' full cooperation. The most valuable knowledge, the kind of nuanced, contextual, and tacit information that separates good from great, is locked inside people.
¡°If top performers feel their jobs are threatened by AI models, they will simply take their best work offline, have private conversations, and take notes on paper rather than feeding their expertise into a system they don't trust,¡± says Li.
Therefore, it is in every company's best interest to create incentives for their people to show up as their best selves. This nurtures a collaborative relationship where expertise is amplified, not just extracted.
If top performers feel their jobs are threatened by AI models, they will simply take their best work offline.
A Premium on Discernment
Instead of threatening people with obsolescence, Li contends that this elevates the importance of uniquely human skills. As AI generates increasingly sophisticated outputs, the most valuable employees will be those who can exercise discernment. ¡°People need to evaluate the AI output and determine if it¡¯s useful or not,¡± says Li.
Business leaders must foster teams that can tell the difference between a good solution and a bad one without blindly implementing flawed, AI-generated recommendations. But how do you properly value and reward a top performer when their most significant contribution is no longer just their individual output, but also the unique data and expertise they provide to make an AI system smarter for everyone else?
As AI generates increasingly sophisticated outputs, the most valuable employees will be those who can exercise discernment.
This challenge goes beyond current performance metrics and strikes at the core of how companies recognize employee value. Solving this compensation puzzle will be a critical next step in building an equitable and collaborative human-machine workforce.
Where Research Meets Reality
Li¡¯s insights are shaped by her dual role as a leading academic and a strategic advisor. As a social scientist, she is driven to understand massive economic shifts through the data that people leave behind as they interact with the world. This perspective led her to join the Â鶹´«Ã½ AI Advisory Council, where she gets what academics can¡¯t access¡ªa real-time view into how AI is adopted across the global economy.
Because Â鶹´«Ã½ has so many clients, it is the place to be if you want to observe how these powerful tools are reshaping finance, HR, and other core business functions in real time.
"Because Â鶹´«Ã½ has so many clients, it is actually the place to be if you want to observe how these powerful tools are reshaping finance, HR, and other core business functions in real time,¡± Li says. This position, at the confluence of academic rigor and real-world application, gives her a unique platform to work with the AI Advisory Council to test her theories at scale and provide data-driven guidance on one of the most significant business transformations of our time.
Danielle Li explored how companies can navigate the human-machine frontier at Â鶹´«Ã½ Rising alongside Kathy Pham, vice president of AI at Â鶹´«Ã½.
Report
C-Suite Global AI Indicator Report: AI Is the Ultimate Level-Up
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.