I will be honest with you, because that is the only way this conversation is worth having.
When AI tools started showing up everywhere in my work and in the rooms I facilitate, my first instinct was not curiosity. It was something closer to the feeling I had during my PhD residencies in Brisbane, that combination of intellectual vertigo and private panic that comes from realizing the world has moved further ahead than you had accounted for. My husband Gordon called my PhD years my Frankenstein Period, and that is not far from how my early months with AI felt either. I was experimenting on myself, making a mess, not always sure what I was building, and quietly hoping no one would notice the seams.
I share that because the leaders I respect most have always been the ones willing to admit when they are in the middle of figuring something out rather than performing a confidence they’ve not yet earned. Naomi Feil, a pioneer in dementia research, once said that we are all born with eighty-eight keys on a piano and most of us spend our lives playing the key of C. AI has been the most confronting invitation I have received in years to play keys I had been avoiding, and I want to talk about what I have learned in the process.
There is a quality I have come to associate with genuinely mature leadership, and it has nothing to do with title or tenure. It is the willingness to step deliberately outside your comfort zone when the moment calls for it, and to invest the time and honest effort required to build a real relationship with something new. The leaders I have watched navigate this AI moment most gracefully are not the youngest people in the room or the most technically sophisticated. They are the ones with enough self-awareness to recognize that staying comfortable is its own form of risk, and having humility means beginning again.
The most useful shift I made was to stop treating AI as a single overwhelming thing to confront and start thinking about it the way I think about a family. Every family has members with different personalities, different strengths, and different roles in the household. Some are the deep thinkers. Some keep everything running smoothly day to day. Some are the researchers who always know where to find the right answer. Some are the creative ones who see the world differently. The key is knowing who does what, and when to call on which member for which kind of work.
Getting to Know the Family
In a healthy family, no single member does everything, and the household runs best when each person is called upon for what they genuinely do well. Your AI family works the same way.
Claude, developed by Anthropic, is the writer and thinker of the family. This is the member you bring in when the work requires sustained concentration, careful language, and genuine depth. Claude excels at long form writing, document creation, leadership communications, nuanced analysis, and working through complex ideas with care and precision. Within the Claude family, Opus is the deep specialist, best suited for sophisticated reasoning and ambitious analytical work. Sonnet is the reliable everyday contributor most leaders will work with most of the time, fast and thoughtful across a wide range of professional tasks. Haiku handles quick light requests where speed matters more than depth. Knowing which Claude to call on preserves your session credits and produces a noticeably better result.
ChatGPT, developed by OpenAI, is the versatile all-rounder of the family, the member who can turn their hand to almost anything and rarely disappoints. Where ChatGPT genuinely shines is in image generation, creative brainstorming, coding assistance, and building out ideas across multiple formats simultaneously. GPT-4o handles complex reasoning, rich content creation, and multimodal work including visual tasks with impressive range. GPT-4o mini is the faster, lighter version for straightforward and high-volume work where efficiency matters more than depth. For leaders whose teams are already using ChatGPT, it is worth understanding which version they are working with and whether it matches what they are trying to accomplish.
Gemini, Google's flagship, is the organizer of the family, the one who keeps the household running because they know where everything lives. Gemini integrates directly into Gmail, Google Docs, Drive, and Meet, working natively inside the tools many leaders already spend most of their day in. It can reference your emails, draft inside your documents, pull from your calendar, and move seamlessly across your existing Google Workspace without requiring you to switch contexts. For teams already embedded in Google's ecosystem, Gemini feels less like a new tool and more like a capable colleague who already knows the filing system.
Perplexity is the researcher of the family, the one you out when you need real answers with receipts. Where other family members draw on their training and knowledge, Perplexity searches the live web in real time and returns sourced, cited answers that can be traced directly back to their origin. For current market intelligence, fact-checked claims, competitor research, or a fast synthesis of what is happening in your industry right now, Perplexity does that job with a precision and transparency the rest of the family cannot match.
You Are the Head of the Household
Here is what I want every leader reading this to hold onto, because it’s the piece that gets lost most often in conversations about AI. This family works for you. You do not work for them.
There is a principle in AI development called Human in the Loop, or HITL, and it’s one of the most important ideas a leader can internalize right now. It means that a human being remains the final decision-maker in every workflow that involves AI. The family generates, suggests, drafts, research, and synthesizes. The leader evaluates, edits, contextualizes, and decides. That chain of command never reverses and keeping it intact is both a practical discipline and a leadership responsibility.
Your AI family will occasionally produce something that sounds authoritative and is factually wrong. They will miss the relational nuance entirely. They will generate something technically competent that you know in your bones is not right for your organization or this moment. Your judgment, your knowledge of the people and the context, that is what catches those gaps and makes the output genuinely useful rather than merely impressive.
Discernment, empathy, and relationships remain yours. Your AI family does not know your organization's history, your team's dynamics, or the trust that has been carefully built over time. The conversations that require a leader to sit with someone in genuine difficulty, to read what is being communicated beneath the words, those moments are centrally human.
The culture that holds an organization together under pressure is built through presence and connection, and no family member, however capable, can replicate it. I have come to think of this through the lens of the 3Ps that run through all my leadership work. Pause, plan, and prepare. Your AI family can accelerate all three phases considerably, but the judgment about what to do with what they surface, and how to bring that into relation to the people around you makes that work entirely and irreplaceably yours.
The Best Decision You Will Make This Year
Pick one family member this week and spend thirty minutes with them on something you are genuinely working on. Notice what they do well and where they fall short. Use your judgment to refine what they produce rather than accepting it as finished. Never accept the first suggestion. Always push for more analysis, more credible references, more ingenious ways to look at problems. Bring what you learn back to your team and make it a conversation rather than a directive. Stay in the loop, stay in command, and let the family do what they are built to do while you do only what you can do.
The human skills that matter most in leadership, the presence, the judgment, the empathy, the courage to step into a difficult room and say something true, are not threatened by your AI family. They are made more visible by them. And the leaders who take the time to genuinely know each family member, to understand their individual strengths, and to stay firmly in command of the household, will be the ones who lead with the greatest clarity and impact in the years ahead. Getting to know your AI family may be the best leadership decision you make this year. I say that from experience, growing pains and all.
If this resonated with you, share it with a leader in your network who is still watching AI from a careful distance. The most generous thing we can do for each other right now is make it a little easier to begin.