Management in the Age of AI: Curiosity beats certainty, speed beats perfection
At the NIN panel "Management in the Age of Artificial Intelligence" in Belgrade's Metropol Palace, Natali Delić (Telekom Srbija) and Darko Marjanović (Things Solver, ASEE Group) made one thing clear: AI won't fix your company-your leadership will. Technology is ready. The real bottleneck is mindset, data, and the pace of change.
The mindset shift: curiosity over security
Transformation is both technical and human. Delić stressed that leaders can't treat everyone the same-people have different experience levels and different tolerance for change. Your job is to make the purpose of change obvious and connect it to strategy.
Older employees often resist but stay resilient. Younger employees adapt fast but can tap out just as fast. Strong leadership balances both: clarity on goals, patience with learning curves, and a bias for action.
New leadership is active, not ceremonial. Move fast, learn constantly, and build environments that bend without breaking. And flip the classroom-leaders should learn from teams, especially recent grads who bring fresh methods and new tools from university.
Data before models
AI adoption lags mostly because company data is messy. Think of data as ingredients and the model as the recipe. Good ingredients, good dish. Bad ingredients, no chef can save it.
If you lead this, start simple and concrete:
- Appoint a data owner for each core process.
- Audit the top 10 data sets used in decision-making.
- Define common terms, fix IDs, remove duplicates, set access rules.
- Track three data-quality metrics monthly: completeness, accuracy, freshness.
Global surveys echo this gap between ambition and readiness. See the latest adoption trends and hurdles in independent research like McKinsey's State of AI.
Build a culture that rewards change
Delić emphasized mandatory ongoing education, clear goals, and a feedback culture. Reward those who drop outdated habits and adopt new ones. Don't just host trainings-show why the change matters, highlight positive examples, and give honest feedback where behavior hasn't shifted.
- Set three measurable outcomes tied to transformation.
- Run monthly feedback loops: what to stop, start, continue.
- Block weekly learning hours for every team.
- Nominate internal champions and pair leaders with junior "reverse mentors."
Use AI to solve real work, not FOMO
Marjanović was blunt: chasing AI because it's trendy is a distraction. Start from a business constraint-cost, speed, accuracy-and apply AI where it directly helps. AI won't change your company by itself; your people will, with the right management.
- Pick two clear use cases: support triage, forecasting, claims review, churn alerts, or field scheduling.
- Define an owner, baseline metrics, and a success target before you build.
- Pilot with a small team and a timebox. Ship in weeks, not quarters.
- Treat AI as a junior analyst: ask for options, check the work, you make the call.
Practical 30-60-90 for managers
- First 30 days: Map your top five processes by cost or delay. Run a quick data health check. Start reverse mentoring with recent grads.
- Next 60 days: Launch one AI assistant pilot tied to a clear KPI. Stand up a data-quality backlog and fix the top blockers.
- Next 90 days: Expand the pilot if the KPI moves. Bake the new workflow into performance goals. Publish a simple "AI Do/Don't" for the org.
Leaders still lead
Marjanović's team trains AI agents-digital employees that can take on well-defined tasks. He treats them like teammates: ask for input, consult on options, decide as a leader. That's the posture to adopt-use the tool, own the decision.
The takeaway: curiosity over certainty, clean data over wishful thinking, and real use cases over hype. Move fast, learn in public, reward adaptation.
Helpful resources
- Independent research on adoption and ROI: McKinsey State of AI
- Want structured learning for your team? Explore AI courses by job role or see the latest AI courses.
Your membership also unlocks:
