Preparing for the AI Knowledge Gap

For experienced managers, the introduction of AI felt like getting new superpowers. All those time-consuming tasks and analyses could be done in seconds instead of hours. The world of possibilities felt limitless. Less experienced managers could get answers to complex questions without doing the heavy lifting. But at what cost?

Consider the person who spends the time to read Shakespeare's Hamlet: they understand the story, the critical analogies, and the underlying tensions, and they apply critical thinking to see beyond the plot line. Now think about the person who reads an AI summary: they understand the general story, but they cannot provide details beyond the surface-level storyline.

For so many executives, the AI drive is all about efficiency and saving labor hours. But in an industry where 71% of hotels have open positions they cannot fill despite active recruiting, according to the American Hotel & Lodging Association, and where annual turnover runs 70-80% while fewer and fewer young people explore a career in hospitality, maybe the focus should be on retention, building institutional knowledge, and creating a talent pipeline. When the cost of replacing a single employee can run from roughly $5,900 to nearly $10,000, according to Cornell's School of Hotel Administration, we should also consider the human impact of AI and how we can leverage it to retain and prepare our teams for the future.

The Efficiency Trap

There is no doubt that AI can improve efficiency by moving tedious tasks such as call centers, room service, and back-of-house functions away from an over-stretched workforce. Advances in revenue management, engineering maintenance programs, and forecasting all provide much more efficient and insightful output. But what happens to the hours saved? How does an inexperienced manager know when a forecast is wrong if they don't know how it is built?

Yes, AI gives managers instant answers to all their queries, but do they understand the underlying structure that enables critical thinking? Having access to answers is creating a generational knowledge gap.

If AI is doing the work, how will emerging managers gain the knowledge, reasoning, and judgment that come from doing it? A manager who can present a forecast but doesn't understand the inputs well enough to question its validity is not set up for success. Senior leaders understand that their success was built on years of analysis, situational awareness, emotional intelligence, and business "smarts" applied to the ever-changing dynamics of everyday life in hospitality. They can sense when things are off. They can look at a budget and know it is wrong. They know how to adjust before things become critical because they have years of applying critical thinking while understanding the business's underlying nuances.

Two Paths

At this juncture, there are two paths to take. The first is to keep AI efficiency as the primary goal and create a streamlined, AI-driven hotel experience modeled to maximize margin. The second is to address the elephant in the room and start deploying AI to bridge the knowledge gap, so managers build the operational knowledge to understand the business and the technical fluency to use AI to manage it more effectively. This can be applied to the onboarding process of line employees and to developing future senior leaders.

The research backs the concern. Microsoft Research and Carnegie Mellon University surveyed 319 knowledge workers and found that the more confident people were in AI, the less critical thinking they applied — raising concerns about long-term overreliance and diminished independent problem-solving. Two of Microsoft's most senior engineering leaders, Mark Russinovich and Scott Hanselman, have described the same dynamic in software teams: agentic assistants give experienced engineers an "AI boost" that multiplies their judgment, while imposing an "AI drag" on early-career developers who lack the judgment to steer and verify the output. Substitute "hotel manager" for "engineer" and the warning applies directly to our industry.

This is not a call to go back to spreadsheets and abandon AI programs; in fact, it is quite the opposite. We have to recognize that there is an apprenticeship for judgment. Yale's K. Sudhir observes that organizations now face a real challenge: junior employees traditionally built skills by doing the exact work that AI now absorbs, and nobody has designed a replacement. Many employees report that they no longer even have access to that work, since AI has taken on those tasks. Organizations need to fill that gap.

The Talent Already on the Payroll

The timing could not matter more, because the people operators need are often already on the payroll. SHL's Talent Mobility at Work in 2026 study, reported by HR Dive in October 2026, surveyed more than 2,000 workers in the US and UK and found that two-thirds expect to leave their jobs in order to advance. Only 42% said their current manager understands their skills beyond their current role, and 59% of US workers believe their employer favors outside hires over internal development. In a time when institutional knowledge is becoming a scarce commodity, developing and retaining the talent we already have is the most rational strategy available.

That development starts at the front door. The first 90 days are a crucial period for any new employee, yet in many cases the onboarding process is broken: new hires receive limited training, frustration grows, and they move on — some to other hotels, some out of the industry altogether. This is an ideal area to deploy AI with retention in mind: onboarding programs that simulate guest interactions, interactive AI-driven training, and new metrics for retention, eNPS, and time-to-fill.

It is crucial that managers build a solid foundation of knowledge. This will require rethinking the entire leadership pipeline and development strategy. Even though AI can handle many of the menial tasks of the past, it doesn't mean new managers get a pass on understanding the components and applying critical thinking. That means programs where managers must present and defend their assumptions, scenario exercises that require them to explain their reasoning, and a formal development program with senior leader advisors. They need to demonstrate that they can apply knowledge, engage in critical thinking, and challenge assumptions rather than just generate prompts for AI and read the output.

What To Do Now

For operators ready to act, three moves address the gap directly:

• Rebuild onboarding around AI: simulated guest interactions, adaptive role-specific training, and retention metrics — 90-day retention, eNPS, and time-to-fill — tracked from day one.

• Institutionalize the apprenticeship: make emerging managers present and defend forecasts, budgets, and schedules, with senior leaders questioning the assumptions, before any AI output is accepted.

• Capture institutional knowledge before it walks out the door: convert veteran expertise into a living, searchable knowledge base that AI delivers at the moment of need.

We can't automate the basic functions that build foundational knowledge and expect leaders to inherit institutional knowledge. In many cases, the work being eliminated is the exact work that helped today's leaders learn and refine their craft. If hotel companies don't deliberately replace that learning, they may find that today's efficiency gains are the leadership deficiencies of tomorrow.

Some will note that hotel turnover has actually declined as AI adoption has advanced — monthly turnover fell from 6.07% in June 2024 to 5.35% in June 2026, according to AHLA and Actabl — and they will be right that the two can coexist. But a turnover rate that remains the worst of any major industry is not a victory; it is a warning about how little slack is left in the system. Aimbridge CEO Craig Smith, whose company is investing heavily in AI-driven forecasting, has built his operating model around a simple principle: no forecasting tool knows a hotel better than the people running it. The tools should multiply that judgment — not replace the years of work it takes to build it.

Don't let your AI strategy become a cognitive debt; turn it into a superpower for future leaders.