Durable Skills Are a Moat in the AI Era
By Lakshmi Nidamarthi, Founder and CEO, Waymakers
When I gave my talk at WACEFE, I looked out at a room of educators and policymakers, and I could feel what they were carrying. Some were anxious. Some were unsure what to tell the students and families who depend on them. Plenty were just curious and hoping someone would say something useful. I understood all three, because I feel them too, and I build AI companies for a living.
Here is what I told them. AI literacy + durable skills = AI readiness. Weeks later, I keep coming back to how well that idea travels. I said it to educators, but it is the same test I apply to founders in the Waymakers Accelerator.
Three terms, translated for people who build companies
AI literacy means knowing what a model can and cannot do. A literate founder can tell when an output is solid and when it only sounds confident.Building with critical thinking skills applied to AI outputs is critical when handling any kind of sensitive data.
Durable skills are judgment, communication, critical thinking, creativity, and adaptability. In a startup, that means staying steady when the product changes every quarter, and explaining a hard tradeoff to a customer, an investor, or a new hire. Using AI does not negate the need to use these skills, it enhances the effectiveness of well developed durable skills. It is a cycle which the founder is always managing.
Readiness is what happens when the first two are present together. It is the difference between a team that is merely building with AI and a team that is fundable and scalable in the coming years.
Why this matters now
The World Economic Forum's list of the ten fastest-growing skills through 2030 has eight that are durable rather than technical. Lightcast's labor market data points the same way: 76% of US job postings over the last two years asked for at least one durable skill, and the five most requested durable skills appeared 3.8 times as often as the five most requested hard skills.
The market is telling us something. As tools get better at producing answers, the scarce thing is knowing which answer to trust, and how to communicate those answers with our audience and customers.
The quiet risk: offloading
There is a pattern I want founders to watch. We already know about cognitive offloading, where people hand their thinking to a tool and slowly lose the habit of doing it themselves. I now think leadership offloading is snowballing. A manager lets a model draft the hard feedback. A founder lets a model summarize the customer calls instead of listening to them. Putting feedback in the hands of AI models can decrease their effectiveness and perceived value.
Education saw this first. Students get frustrated when a professor's feedback is clearly written by AI. They can tell no one was paying attention. Workplaces work the same way. If your team senses that judgment is being outsourced, trust erodes, and culture goes with it.
None of this is an argument against using AI. Our whole portfolio is built on it. It is an argument for deciding on purpose where a human stays in charge. We believe keeping humans in the loop is a key part of making companies successful in the AI era. It is not too late to keep the trust we built before AI models were introduced to the consumer world.
How this shows up in how we pick founders
When we screen a company, we score it against a written rubric. Leadership and product-market fit each carry the highest weight in the rubric. AI-nativeness carries the same weight, so technical depth matters, but it does not outrank the other two. Mission, unit economics, sector fit, and funding stage are important, but not at the same level as the founding team, and whether the product can actually be useful to the customer.
Both of the heaviest human factors are made of durable skills. A strong founding team is a judgment story. Product-market fit is a listening story, because it shows whether a founder heard what customers were really saying.
The founders who make it through are not only the ones who can build with AI. They can reason about what they built. They catch the failure before a customer does. They know when a fast answer from a model is the wrong answer.
These habits are also built into how we work, not just how we talk. Our operators work inside a company for 9 to 12 months, and much of that time goes to the decisions that require a person: what to say no to, how to handle a bad quarter, when to slow down and check on the broader market conditions.
What I would tell you if you are anxious
If you are worried about where this is going, I would not tell you to worry less. I would tell you to go back to the basics. Get curious about what the tools actually do. Practice the skills that were valuable before AI and are more valuable now. Notice when you are handing off a decision you should be making.
That is not a soft skill. It is the moat.
This is the first in a series on durable skills. Next, we look at what they mean for parents and caregivers, and how to measure them at work. Subscribe to follow along. If you are building an AI company in health, education, or workforce, you can apply to Waymakers at waymakers.ai.
AI tools were used to help with editing, research, and linking. The ideas and arguments are Lakshmi's.