Working With AI: Seven Practices of Collaborative Inquiry

Working with artificial intelligence is not about delegating thinking to a machine. It is about creating a critical and reflective process through which ideas can be explored, assumptions examined, perspectives expanded, and possibilities co-imagined before decisions are made.

Collaborating with the AI. Original design by Andres Siimon

As artificial intelligence becomes increasingly integrated into everyday life, a growing number of people are asking a practical question: How should we work with it?

Much of the advice currently available focuses on prompts, productivity, and efficiency. We learn how to automate tasks, generate content more quickly, or obtain better results from AI systems. These skills are undoubtedly useful. Yet they only address part of the picture. The more interesting question may not be how to get more from AI, but how to think, learn, and create with it.

In a previous article, we suggested that artificial intelligence can be approached not only as a tool or a potential competitor, but also as a partner in processes of thinking and learning. This perspective shifts attention away from questions of replacement and control toward questions of collaboration.

Collaboration, however, is more than interaction. We collaborate when we do not merely exchange information but actively shape a process and its outcomes together. In this sense, collaboration needs meaningful participation, which is more than being present or having a voice. It involves contributing perspectives, helping define directions, and taking part in the decisions through which outcomes emerge.

This is why collaboration is also a question of power-sharing. It requires some degree of shared responsibility and shared influence over a process. Rather than concentrating control in a single actor, it opens possibilities for what we described in the previous article as "power with"—the capacity to think, learn, and create together.

While human-AI relationships differ in important ways from human-to-human collaboration, this insight remains useful. The quality of collaboration depends not on how many prompts we write or how many answers we receive, but on how AI and us both become part of processes of inquiry, reflection, interpretation, and decision-making.

Working with AI, in this sense, is not about delegating thinking to a machine. It is about creating a more critical and reflective process through which ideas can be explored, assumptions examined, perspectives expanded, and possibilities co-imagined before decisions are made. While there is no single way to collaborate with AI, certain practices can help cultivate more thoughtful, reflective, and meaningful forms of engagement.

The following seven practices offer one potential pathway.

1. Explore Possibilities, Not Just Answers

Many people turn to artificial intelligence looking for answers. Yet its greatest value often lies elsewhere: helping us see possibilities we might not have considered.

When developing a project, exploring a research question, or responding to a complex challenge, it can be tempting to ask for the best solution immediately. But meaningful inquiry often begins before solutions emerge. It begins by expanding the space of possibilities.

Instead of asking "What should I do?", try asking "What possibilities am I not seeing?" or "What alternative ways of approaching this issue exist?" AI can help surface unexpected connections, unconventional ideas, and paths that may otherwise remain invisible. In this sense, its role is less about providing certainty and more about widening the horizon of what seems possible.

2. Surface Hidden Assumptions

Every argument, project, or decision rests on assumptions. The difficulty is that we are often unaware of our own.

One of the most useful ways to engage with AI is to use it as a mirror for reflection. Ask it to identify assumptions underlying your thinking, challenge your reasoning, or point out perspectives that may be missing from your analysis.

This practice is particularly valuable because many of the limitations we encounter are not caused by a lack of information but by the boundaries of our existing mental models. Sometimes progress begins not with finding better answers, but with questioning the assumptions that shape the questions themselves.

3. Practice Perspective-Taking

Complex issues rarely look the same from different vantage points. Whether we are discussing education, technology, climate change, migration, or community development, different groups often experience the same reality in very different ways. AI can help us temporarily step outside our habitual frames of reference and explore alternative viewpoints.

How might a teacher approach this issue? A policymaker? A community organizer? A young person? Someone from a different cultural context?

The purpose is not to determine which perspective is correct. Rather, it is to cultivate the capacity to engage complexity with greater curiosity and empathy. In an increasingly polarized world, the ability to understand multiple perspectives may be one of the most valuable forms of intelligence we can develop.

While these first three practices focus primarily on expanding our own thinking, inquiry rarely remains an individual activity. Understanding complex issues often requires engaging perspectives beyond our own.

4. Think Across Boundaries

Some of the most interesting ideas emerge at the intersection of different fields of knowledge. Yet educational systems, professional environments, and academic disciplines often encourage us to think within established boundaries. AI can serve as a useful companion in crossing those boundaries and exploring unfamiliar intellectual territory.

What might ecological thinking contribute to leadership? What can artists teach us about conflict transformation? What might communication studies offer to conversations about artificial intelligence?

Such questions rarely produce immediate solutions. Their value lies in generating unexpected connections and opening new avenues of inquiry. Innovation often begins where previously separate worlds start to meet.

5. Engage in Dialogue, Not Delegation

Perhaps the most common mistake in using AI is treating it primarily as a system that performs tasks on our behalf. While delegation certainly has its place, collaborative inquiry requires something different. It requires dialogue.

Rather than asking AI to produce a finished article, report, or presentation, invite it into an ongoing conversation. Ask it to challenge your argument, offer alternative interpretations, or help refine an emerging idea. Explore points of disagreement. Follow unexpected threads.

In working with the AI, the goal is not to outsource thinking but to create a dynamic process through which ideas can evolve. Some of the most valuable insights emerge not from receiving answers, but from sustained dialogue.

6. Learn Through Questions

Many educational systems reward the ability to produce correct answers. Yet transformative learning often begins with the cultivation of better questions.

This is one of the most overlooked ways of working with AI.

Instead of asking for immediate explanations, try asking AI to help identify the most important questions surrounding a topic. What remains uncertain? What deserves further investigation? What assumptions should be examined more carefully?

The quality of our learning is often shaped less by the answers we receive than by the questions we learn to ask. In this sense, AI can become less a source of information and more a companion in inquiry and discovery.

Inquiry also extends beyond understanding the present. It invites us to consider the futures our actions, assumptions, and choices help bring into being.

7. Co-Imagine Possible Futures

Perhaps the most exciting use of AI is not solving today's problems but exploring tomorrow's possibilities.

AI can help us imagine alternative futures, examine different scenarios, and reflect on the long-term implications of choices made today. It can support conversations about what kinds of societies, institutions, communities, and relationships we wish to cultivate.

Which futures become possible under certain conditions? Which voices are included and which are excluded? What assumptions about progress, wellbeing, or success shape our visions of the future?

These questions move us beyond efficiency and productivity toward something more fundamental: our collective capacity to imagine, discuss, and shape desirable futures. In a world facing profound ecological, technological, and social transformations, this may be one of the most important forms of collaboration that AI can support.

Practices in Practice

The practices elaborated above are not separate techniques to be applied in isolation. They can be understood as a work flow. They begin with individual reflection and the expansion of one's own thinking. They then move toward perspective-taking, dialogue, and moving beyond boundaries. Finally, they extend toward collective questions about the futures we are creating together. In practice, these dimensions often overlap. A good question may reveal hidden assumptions. A dialogue may open new possibilities. Exploring alternative futures may require engaging multiple perspectives. What matters is not following a sequence, but cultivating a more reflective and collaborative way of working with AI.

Yet there is another important point. While these practices often begin with individual inquiry, they rarely end there. Most of the challenges we face today—from ecological crises and social polarization to technological transformation and democratic participation—cannot be understood or addressed by individuals alone.

Inquiry is rarely a solitary activity. It unfolds in classrooms, research teams, communities, organizations, and networks of collaboration. This raises a broader question: What happens when AI becomes part of collective learning processes rather than individual ones?

In the next article, we explore how AI might contribute not only to individual reflection and creativity, but also to collective intelligence, shared learning, and the emergence of new learning ecologies.