How Leaders are Leveraging AI to Put the Lid on Brain Drain

Organizations are currently struggling with employee retention, and some professionals may even opt to leave one region for another in search of better career opportunities.According to SHRM’s Global Employee Monitor, the most influential factor in retention for Q2 was pay — 46% of employees reported having left a job due to this factor. With each employee departure, companies run the risk of losing institutional knowledge.

To combat the potential impacts of what is often referred to as “brain drain,” some companies are using artificial intelligence to capture and share knowledge. At ERIA, a hospitality and events company in northern California, the team is training ChatGPT and Claude to help protect the organization from the impacts of attrition and retirement by creating bots that have the institutional knowledge. 

Other examples of AI for knowledge capture include Hitachi’s Generative AI Center and Abb’s Industrial Knowledge Vault on Microsoft Azure’s OpenAI Service. 

“I think companies make a mistake when they wait until someone is leaving to think about knowledge transfer. By then, you’re trying to download years of judgment in two weeks,” said Nikita Khandheria, founder and CEO of ERIA. “[The goal is] turning knowledge that traditionally lives in my head or with experienced team members into institutional knowledge the company can keep.”

The Knowledge Transfer Problem

Retirements, buyouts, and organizational flattening are just a few of the factors that are destroying decades of institutional knowledge in a business environment where remaining nimble and competitive are must-have capabilities. 

While the conventional response to knowledge management has evolved around knowledge management systems and structured documentation, these efforts have largely missed the mark because they’re not focused on understanding the “why”.

“AI is most useful against brain drain when it captures why experienced employees made decisions, not just what they did,” said Gleb Tsipursky, PhD, behavioral scientist and author. “I would focus on using AI to capture context before knowledge disappears, rather than merely archiving documents.”

Tsipursky advises considering questions such as:

• When does the normal process fail? 
• Which signals tell you that?

Specialty Credential: AI+HI

Success Stories of Leveraging AI

When used correctly, AI can help employees make better informed decisions. According to Khandheria, traditional employee handbooks struggle to capture what’s behind common workplace decisions, but AI can turn these situations into interactive training. 

At ERIA, the team has created training using AI for increased engagement. “Team members can work through scenarios, ask questions and understand not just what ERIA does, but how we think about a problem and why,” she said.

AI can also be used to summarize institutional knowledge into documentation. At consultant Momentum Group, the team recorded conversations about real scenarios, decisions, and exceptions into an AI tool which then created training documents. The recordings captured how experienced employees actually navigated systems and the undocumented shortcuts that had been developed.

Jackie Cook, founder and CEO at Momentum Group cautions leaders to ensure that AI structures rather than validates the knowledge. “The people actually doing the work still need to pressure-test the output against reality," she said. 

AI can also be used to help answer HR questions, shortening queue wait times to a matter of seconds. “[HR questions used to take up to five days to resolve, because the answers lived with a small number of people who became a queue,” said Surojit Chatterjee, CEO of Ema, an agentic AI platform.

Toolkit: AI Field Manual

What CHROs Should Do Now 

Below is Chatterjee’s advice to CHROs hoping to collect institutional knowledge before it’s lost through attrition.

• Stop treating knowledge capture as an event. “A documentation sprint ahead of a retirement wave will fail the same way every previous one did. Capture has to be continuous and passive," he said, " a byproduct of daily work, not a task added to it.”
• Start where knowledge density and attrition risk overlap. “Payroll exception handling, benefits administration, compliance interpretation, and customer escalations are the usual concentration points. Map tenure against function and the priority list writes itself.”
• Measure four things: time-to-answer for employee questions, time-to-productivity for new hires in the target function, escalation rate to named experts, and percentage of queries resolved without a human specialist. “[These metrics] can move within a quarter, and they are numbers a CFO will accept.”
• Take governance into consideration at the outset, not after deployment. Institutional knowledge includes compensation data, investigations, and health information. “Any system surfacing it must inherit your existing access controls, audit trails, and privacy obligations,” he says. “Ask vendors to show role-based permissions and audit logs in the demo, not in the security review three months later.”
• Tell the workforce why you’re doing what you’re doing. In an environment of buyouts and flattening, an initiative to "capture what employees know" can be perceived as replacement planning unless leadership says otherwise, he warned. “The organizations that succeed frame it honestly: we’re making your expertise reusable, and we’re removing you as the bottleneck for questions you're tired of answering.” 

The organizations that successfully battle brain drain aren't the ones buying the most sophisticated AI tools. They're the ones redesigning how knowledge is captured, governed,and shared — all while the knowledge-holders are still there to explain.