Win your Future in the Agentic AI Era

 Cisco’s former CEO John Chambers said that reinventing not only your company but yourself is the critical skill for every leader in the digital age. That has rarely been truer. AI has moved from chatbots that answer questions to agents that plan, use tools and finish multi-step work. Change is constant, and the pressure to keep up is real.

Every era of rapid change feels unprecedented until it becomes normal. Here are 20 keys to staying in charge of your career through this one, each with something you can try this week.

Part I: Work with agents

1. Direct agents, don’t just chat with them. The skill that matters now is briefing, decomposing and reviewing. Google’s Antigravity 2.0, for instance, is built to orchestrate multiple agents in parallel, with one coding a website while another generates brand assets. Treat agents like a fast, literal junior team. Try this: write a one-page brief for a task you’d give a new hire, hand it to an agent, then critique the output as you would theirs.

2. Delegate what you can verify. Anthropic’s report on agentic coding finds engineers tend to hand off easily checked or low-stakes tasks and keep design-heavy work for themselves. Caution is warranted: in Stack Overflow’s 2025 survey, only 29% of developers trusted AI output to be accurate, down from 40% the year before. Try this: before delegating, ask “how would I know if this is wrong?” If you can’t answer, don’t delegate yet.

3. Double down on human strengths. As drafting, boilerplate and summarizing get automated, the premium shifts to judgment, ethical reasoning, empathy, and the ability to build trust and move people. Communication, persuasion and selling will only grow in value. Reviewing well is now part of the job.

4. Cultivate taste and originality. AI produces endless first drafts, so the scarce skill is knowing which draft is good and why. Creativity is trainable: keep a notebook, protect solitude, move your body, and talk to people who make things. Try this: each week, make one thing (a memo, a sketch, a prototype) before asking AI for help.

5. Learn continuously, and skeptically. Attention is scarcer than information, so pick sources that compound your judgment and ignore the rest. Be careful with headline stats. Google says 75% of its new code is AI-generated and approved by engineers, but the definition behind that figure has changed several times in 18 months. Read the definition before you believe the number. Kodak invented the digital camera and still failed to reinvent its business, so keep watching the shifts that could disrupt you. Try this: fix one daily learning slot and keep a short “not learning” list.

Part II: Own your career

6. Find work that matters. You can bring passion to the work you already do. Hubert Joly asks in The Heart of Business: what cathedral are you building? People who care about outcomes are harder to replace, and meaning is a competitive advantage.

7. Manage yourself. Peter Drucker argued that the defining challenge of our time is managing ourselves, because so many people now have real choices. Know your strengths and focus on contribution. Try this: log where your time goes for one week, then protect your best hours for your most important work.

8. Build a clear USP and a visible brand. Be known for something specific. When people hear your name, they should link it to a particular strength. Learning new skills keeps you employable, but visibility makes you findable. Try this: write one sentence, “I help ___ do ___,” and publish something that proves it every month. You are a brand, build it and be known by your USP.

9. Cultivate self-reliance. Be the CEO of your own career. Andy Grove’s Only the Paranoid Survive is about spotting strategic inflection points before they hit, and each of us has our own. Try this: keep a “proof of work” file of results you can point to, and update it monthly. You are the CEO of your career.

10. Become a capable generalist. As I have mentioned in my other articles David Epstein’s Range argues that breadth and the ability to connect ideas across fields create advantage, as Roger Federer’s varied athletic background illustrates. In an agent era, orchestrating specialists, human and machine, often beats deep expertise in one narrow skill.

11. Practice creative abandonment. Ask regularly: knowing what I know now, would I start this again? Jack Welch pushed GE to stay only in businesses where it could be number one or two. Ignore sunk costs, including skills and tasks that agents now do better. A clear business example is Microsoft under Satya Nadella. After years of heavy investment in Windows Phone and other legacy priorities, the company deliberately walked away from those bets to concentrate on cloud and AI. The decision freed capital, talent, and leadership attention for the areas that ultimately drove its resurgence. Try this: each quarter, write a “stop doing” list.

Part III: Grow with others

12. Be a leader. Set direction, resolve ambiguity, model optimism, share credit and treat people well. As agents multiply, the humans who align teams around a goal become the force multipliers.

13. Welcome feedback. Everyone has blind spots. Ask bosses, mentors and peers, then filter what you hear against what you know about yourself. Marshall Goldsmith’s What Got You Here Won’t Get You There and his feedforward technique are practical tools. Try this: ask three people, “What’s one thing I could do differently?”

14. Embrace vulnerability and drop perfectionism. Brené Brown’s work makes the case for openness in leadership. Ship early and improve. The Minimum Viable Product idea from The Lean Startup applies to careers as much as products, and it includes having hard conversations cleanly. Dropbox provides a classic illustration. Instead of building a fully polished product first, the founders released a simple three-minute demo video that showed the core idea. The imperfect, incomplete version generated massive early interest and validated demand long before the product was finished. Many teams that wait for perfection never learn whether the idea is worth pursuing. Try this: ship one imperfect version of something this month and collect real feedback.

Part IV: Stay steady under uncertainty

15. Stay resilient. Resilience is how you respond to setbacks you didn’t choose. Anchor yourself in stable core beliefs, extract the lesson, and replace worry with action. Angela Duckworth’s work on grit is a good companion. Try this: after a setback, write a short note: what happened, what I control, what I’ll do next.

16. Stay flexible. Flexibility is changing course by choice. Yesterday’s playbook can become today’s liability. General Eric Shinseki put it well: if you dislike change, you’ll dislike irrelevance even more. Try this: set tripwires in advance, such as “if X happens, I’ll reconsider Y.”

17. Choose grounded optimism. Optimism without realism is delusion, and realism without optimism is paralysis. Try this: name the realistic worst case, then write the next action you’d take.

Part V: Protect your energy and attention

18. Protect your attention. Strip non-essential notifications and apps from your phone, and use social platforms through a browser. Focus is a superpower when everyone else is fragmented.

19. Manage your energy. Sleep, movement, nutrition and stress management set your capacity. Journaling, a habit Benjamin Franklin practiced, remains one of the simplest ways to process emotions and keep clear.

20. Work in sprints and recover deliberately. Going all-out without recovery brings diminishing returns. Schedule renewal weekly, whether reading, exercise, travel or something else that restores you. Rest is a performance strategy.

Reinvention isn’t a one-time event but a practice. Those who treat it that way won’t just survive this shift, they’ll help shape what comes next. You can win in the Age of AI with these strategies. Get to your best and be your best.

The views expressed here are my own and do not represent my organization.