Five human skills that grow more valuable with AI
My son asked why he should learn programming if AI will do it for him. The skills that matter most are becoming more valuable: critical thinking, empathy, language, value creation, judgment and taste.
Written by
-
Benedikt Stemmildt Founder & Co-CEO
Published: 7 October 2025 Updated: 2 min read
My son asked me at dinner last week: “Dad, why should I learn programming if AI will just do it for me?”
His question hit differently than I expected. He’s right to ask, and wrong about the conclusion.
Here’s what I’ve learned from teaching, now in the AI era: the skills that matter most are becoming more valuable, not less.
1. Critical thinking: evaluating the path, not just the destination
The world needs people who can critically evaluate how the AI arrived at its solution.
Not blindly accepting, not blindly rejecting: being skeptical and stoic. Asking what assumptions it made, what alternatives exist, and why it took this path.
AI can optimize its own process. Only humans decide if we should trust that optimization.
2. Deep empathy: knowing what actually matters
AI can discover problems and figure out how to solve them. But AI can’t know if those problems matter on a human level: to your family, to your friends, to your community.
Which solutions create joy, real connection, a real improvement? That requires understanding humans at a level AI can analyze but never truly feel.
3. Language proficiency: the foundation AI learns from
Here’s the irony: AI is a large language model. Its power is language.
The better you are with language (reading, writing, logic, clarity), the better you’ll work with AI. People who struggle with AI aren’t those who can’t code. They’re those who can’t communicate clearly what they want.
4. Authentic value creation: being a partner worth having
When technical execution becomes a commodity, what sets you apart?
Being someone who cares about creating value for others. Being someone people want to team up with. Being someone who makes the team better.
AI won’t replace developers. But developers who create authentic value with AI will replace those who don’t.
5. Judgment and taste: what the algorithm can’t replicate
AI can show you 10 ways to solve a problem. Only you decide which one aligns with your values, your team culture and your product vision.
Taste isn’t arbitrary preference. It’s judgment developed from experience, empathy, and understanding trade-offs. It’s the human strength AI amplifies but never replaces.
What I tell young developers now
“Learn to program, yes. But more importantly: learn to think critically about what you build. Learn to understand people deeply. Learn to communicate with clarity. Learn to create real value. Learn to make wise decisions.”
These skills don’t just survive AI. They thrive with it. The future isn’t about competing with AI. It’s about being so good at these human skills that AI multiplies them.
Which of these skills are you actively developing?
Further reading
Written by
-
Benedikt Stemmildt · Founder & Co-CEO
More on Adoption
- Three windows for adopting agentic coding are closing Teams that plan to adopt agentic coding in two or three years underestimate the learning curve, the talent market and the economics. Each of the three gives a reason to start the clock now.
- Engineers finally get to work on the system When an agent gets it wrong, the fix is a rule, a skill or a hook rather than an edit to the code, so the engineer's job moves from working inside the system to working on it. Engineers were rarely given that job and have few criteria for doing it well, so a team has to decide what the work is judged by.
- Give every team one AI hypothesis instead of generic training Generic AI training across 300 people produces compliance, not adoption. One written hypothesis per team, tested on a retro cadence, makes the rollout correct itself.
Test the constraint in your system
The Factory Readiness Assessment connects this research to your value stream, your data, and the next decision.
Training on this topic: Agentic Engineering for Engineering and Product LeadersWant to check one team first? The self-check: twelve yes or no questions, three minutes, instant result, no account. Take the self-check