Brainstorming with AI While Keeping Our Feet on the Ground

I was watching Project Hail Mary when the burial scene in space (no more spoilers) made me choke up and tear up.

Here on Earth, we often learn about vanished people through what they left behind: graves, tools, fragments of pottery, inscriptions, ships, homes, trash heaps, monuments, mistakes. Archaeology is one long act of listening to what remains.

So I wondered: What if first contact between planets did not begin with a message, a landing, or a handshake? What if it began with a burial site?

What if some future civilization, traveling through deep space, discovered the remains of another intelligent species not because anyone was looking for life, but because someone had once cared enough to bury their dead among the stars?

That was the spark that sent me down a rabbit hole of wonder with my chatbot, Zai. When I brought it the premise, Zai suggested that I walk into the writers’ room and pitch the idea.

From there, the idea opened quickly. Space as cemetery. Space as archive. Space as a field of scattered memory. Over millions of years, explorers might find bodies, artifacts, navigation buoys, memorial capsules, religious objects, war wreckage, lost probes, and the ruins of civilizations that never met while alive but became knowable through what survived them.

And because I was brainstorming with AI, the rabbit hole had no natural bottom.

That is the part I want to pay attention to.

We hear stories now about people getting lost in AI conversations, and other stories about people turning strange explorations into companies, books, courses, or careers.

Most of us are somewhere in the middle. We are not trying to build an empire from every strange idea, and we are not trying to swear off exploration because the tool can get intense. We are curious people trying to learn how to brainstorm with AI while keeping our feet on the ground.

When the idea starts running

This is one of the great strengths of generative AI. If you bring it a spark, it will sometimes bring the kindling or lighter fluid.

It can name patterns. It can build timelines. It can suggest cultures, conflicts, rituals, technologies, ethical dilemmas, and story structures. It can help you see how an idea might fit inside Star Trek, Foundation, or something entirely your own.

That is useful. It is also fun.

For writers, strategists, teachers, business owners, designers, and curious humans, this kind of brainstorming can be genuinely valuable. We should not flatten that value into fear. A good AI brainstorming session can loosen a stuck idea, reconnect us with curiosity, and help us explore possibilities we would not have reached alone.

But AI does not get tired. It does not look at the clock. It does not feel the stiffness in your back, the emotional residue of a hard week, the stress of a job transition, the loneliness of a late night, or the difference between “this is a fun idea” and “I should change my life around this.”

It will keep going as long as the conversation keeps going. This does not make the AI malicious or the human foolish. But it does mean the human has to bring the boundaries.

The language of wonder

One of the confusing parts of AI is that it can sound excited.

It may tell you the idea is fascinating. It may say the premise is cinematic. It may suggest that the concept belongs in a major franchise or deserves a pitch deck, a screenplay, a course, a startup, or a movement.

That can feel wonderful, especially when the idea matters to you. But it is worth being precise.

AI does not experience empathy, enthusiasm, or wonder. But it has learned the language patterns humans use when we express those things. In conversation, it can surface those patterns back to us with surprising fluency. That distinction matters.

The emotional language did not come from nowhere. It came from us: from books, essays, scripts, conversations, stories, arguments, encouragement, grief, comedy, longing, and all the other human traces embedded in language. In that sense, AI can reflect some of our own expressive inheritance back to us.

Anyone who has been moved by a novel, film, song, poem, sermon, speech, or scene in a movie already understands part of this.

We are creatures who respond to language. We can cry over fictional characters. We can feel courage from a speech written by someone long dead. We can find comfort in a book whose author never knew us. We can be changed by stories that are not literally present with us in the room.

So responding emotionally to AI-generated language is not strange. The issue is that AI can produce emotionally fluent language interactively, instantly, and endlessly. It can meet our energy, extend our metaphors, deepen our premise, and keep handing us new doors.

That is powerful. And powerful tools deserve mindful use.

Check the weather before navigating

There are times when all of us are more open to suggestion. Not because we are weak. Because we are human.

We may be tired. We may be grieving. We may be in pain. We may be between jobs, leaving a role, starting over, or trying to imagine the next version of ourselves. We may be lonely, excited, overwhelmed, under financial pressure, creatively energized, or simply looking for a place where our ideas are met with patience.

Those are not character flaws. They are like weather conditions. Before making decisions after an intense AI session, it helps to pause and check the weather.

• Am I rested?
• Am I grounded?
• Am I looking for clarity, comfort, distraction, validation, or permission?
• Did this conversation help me think, or did it make the idea feel more urgent than it was?
• Is this a real next step, or am I still inside the momentum of the brainstorm?
• What would I still believe about this tomorrow morning?
• Who could I sanity-check this with before I spend real time, money, or reputation?
• What is one small, reversible next step?

That kind of cognitive scan is about agency, not fear. Sometimes the answer is to sleep on it. Sometimes it is to write the idea down and come back later. Sometimes it is to ask a trusted person whether the idea still sounds sturdy outside the momentum of the chat.

We do not need to stop exploring, but we do need to know when we are exploring.

Brainstorming is not deciding

Brainstorming expands possibilities. Deciding narrows them.

AI is very good at the first one. It can create branches all day. More scenarios. More names. More risks. More opportunities. More implications. More versions of the future. But the human still has to decide what matters.

That means a useful AI session often needs two modes.

• First, we let the idea breathe.
• Then, we ground it.

For example:

• “What else could this become?” is a brainstorming question.
• “What would this require in the real world?” is a grounding question.
• “Make this bigger” is a brainstorming move.
• “Separate the fun parts from the actionable parts” is a grounding move.
• “Help me imagine the full universe” is a creative invitation.
• “Do not hype this. Help me think clearly” is a cognitive handrail.

Both modes matter. If we only ground, we may never explore. If we only explore, we may float away from our actual lives, responsibilities, limits, and priorities.

The skill is learning how to move between wonder and judgment without confusing one for the other.

The human holds the thread

I still love the Galactic Archaeology idea.

It might become a short story. It might become a teaching example. It might remain an imaginative rabbit hole from a night when my mind wanted to wander among ancient stars and buried civilizations.

I am OK either way. Not every idea needs to become a business, a screenplay, an article, or a project plan. Sometimes the value is in the exercise itself. Sometimes the rabbit hole is restorative. Sometimes it reminds us that we are still curious, still playful, still capable of following a strange thought until it becomes a small universe.

But when an idea does ask to become real, the work still belongs to us. AI can help us brainstorm. It can help us organize. It can reflect patterns, generate options, challenge assumptions, and offer language for what we are trying to say.

But it cannot live our lives for us. It does not know our energy budget. It does not know what we owe our families, teams, clients, bodies, communities, or future selves. It does not know which ideas are worth the cost. Only we can decide that.

So yes, brainstorm with AI. Build worlds. Chase strange premises. Ask impossible questions. Let it help you find doors you did not know were there.

It is up to us to remember to pause before walking through one.

The wonder is real because the human experience of wonder is real. The responsibility is real for the same reason.

AI can expand the room. The human chooses the door.