Simulated Brainstorms — FAQ

What is a Simulated Brainstorm?

A Simulated Brainstorm is an AI-driven deliberation environment where simulated personas discuss a topic on your behalf. Instead of giving you a single AI answer, it generates a multi-perspective debate — AI personas challenge each other, weigh tradeoffs, and explore risks before arriving at justified decisions. Once the simulation is complete, it converts into a regular Brainstorm where your human team can take over.

How is it different from a regular Brainstorm?

A regular Brainstorm is a collaborative workspace where real team members discuss, share ideas, and make decisions together. A Simulated Brainstorm does the initial exploration with AI personas first — debating the topic, surfacing blind spots, and logging decisions — and then hands off to your human team. Think of it as a head start. Your team picks up from an informed starting point rather than a blank page.

How do I create one?

Click "New Brainstorm" from the Brainstorms dashboard. Fill in the name, description, and goals, then toggle on Simulated Brainstorm. Add members, assign an admin, and optionally add Artifacts (documents, transcripts, data) for context. The simulation will run once setup is complete.

Who are the AI personas based on?

The AI personas are informed by the Wethos Styles of the real team members you add to the Brainstorm. If you add someone with an Ideas 5 and Order 2, the simulation will include a persona that thinks and debates the way someone with that behavioral profile would. This means the simulation reflects the actual cognitive diversity of your chosen team.

Does the simulation use real team members' names?

Yes. The simulated responses are attributed to the team members you've added, but each message is clearly labeled with a Simulated tag so there is no confusion about whether a response came from a real person or the AI.

Can I add context or documents for the simulation to reference?

Yes. After creating the Brainstorm, you can add Artifacts — documents, meeting transcripts, reports, or any relevant materials. The AI personas reference these Artifacts during the deliberation, grounding the discussion in real information rather than generic reasoning. The more relevant context you provide, the more useful the simulation will be.

How long does a simulation take to run?

Simulations typically complete within a few minutes, depending on the complexity of the topic and the number of participants. You can watch the deliberation unfold in real time as the AI personas post messages in the conversation thread.

What are "Decisions" in a Simulated Brainstorm?

Decisions are key conclusions that emerge from the deliberation. They are captured in two ways: XO automatically generates synthesized decision summaries with key metrics (actionable items, source breakdown, contributors, and discussions merged), and specific messages can be manually marked as decisions with a green indicator. Each decision reflects how the differing perspectives were considered and resolved — this is what makes it "justified convergence."

Can I unmark a decision?

Yes. Any message that has been marked as a decision can be toggled with the Unmark decision button if it needs to be revised or reconsidered.

What happens after the simulation finishes?

The Simulated Brainstorm automatically converts into a regular Brainstorm workspace. All simulated conversation history is preserved, all Decisions are logged in the Decisions Hub, and all Artifacts remain attached and queryable through XO. Your human team can then enter the workspace, review what the AI explored, and begin their own collaboration from where the simulation left off.

Can real team members participate during the simulation?

The simulation is designed as an AI-driven phase. Once it concludes and transitions to a regular Brainstorm, human team members can jump in, post messages, tag XO, create Breakouts, and collaborate as they would in any Brainstorm.

Does the simulation replace human decision-making?

No. Simulated Brainstorms are designed to accelerate and inform human decision-making, not substitute for it. The simulation explores perspectives, maps tradeoffs, and surfaces blind spots — but the final decisions and actions are always made by your human team.

How should I choose which team members to add?

Choose thoughtfully. Since the AI personas are based on the behavioral profiles of the members you add, the behavioral diversity of your selection directly shapes the range of perspectives the simulation explores. A team heavy on Visionary thinkers will produce a very different simulation than one weighted toward Methodical executors. For the richest output, include members with a variety of styles.

What should I write in the goals and description?

Be specific. The description and goals fields provide the context that guides the AI personas' discussion. "Explore options for Q3 launch" will produce a less useful simulation than "Evaluate three go-to-market strategies for the Q3 product launch, weighing cost, timeline, and team capacity." The clearer your inputs, the more focused and actionable the simulation will be.

Can I run a Simulated Brainstorm from a meeting?

Yes. You can launch a Simulated Brainstorm directly from a meeting context by injecting meeting-related Artifacts (agendas, transcripts, notes) upfront. This is useful for pre-exploring topics before a live session.

Can I create Breakouts inside a Simulated Brainstorm?

Breakouts become available after the simulation transitions into a regular Brainstorm. At that point, your human team can create Breakouts for sub-topics that need deeper exploration, just like in any standard Brainstorm.

What's the best use case for a Simulated Brainstorm?

Simulated Brainstorms are especially powerful for exploring strategic options before a leadership decision, preparing for high-stakes meetings by anticipating objections, testing a proposed plan against diverse behavioral perspectives, accelerating early-stage brainstorming with a rich starting discussion, training teams on constructive disagreement by modeling how AI personas debate, and pre-exploring complex topics before a live session to save meeting time.

I ran a simulation but the output wasn't useful. What should I do?

The quality of a simulation is directly tied to the quality of its inputs. Try adding more detailed goals and a clearer description, uploading relevant Artifacts so the AI has real data to work with, and including members with a wider range of Wethos Styles for more diverse perspectives. If issues persist, reach out to support.