How WethosAI works

WethosAI is the Human Context Platform. It models the specific people, relationships and organizational context involved in a consequential decision, so you can rehearse how those people might respond before you act. This page explains what that means in practice and how the parts of the product connect. Every other article in this Help Center covers one part in more detail.

Who uses WethosAI, and for what

Organizations use WethosAI wherever the outcome of a decision depends on how specific people respond. The four main uses are:

Cybersecurity, risk and fraud. Run the same incident or fraud scenario repeatedly against the people responsible for it, and test the controls that depend on a person: verification, approval, escalation and crisis response. See Human Control Validation and WethosAI for cybersecurity.
Executive decisions and change. Reorganizations, mergers and acquisitions, leadership transitions and strategic decisions, where the plan only works if the people who approve, influence and carry it out are aligned. See WethosAI for enterprise.
Human + AI workforce design. Decide how people and AI agents work together: who holds which decision rights, who oversees an agent, and when a person must step in or escalate.
Leadership development, coaching and people teams. Prepare for a difficult conversation, a first leadership role or a hard piece of collaboration by practising it first.

WethosAI does not rate or rank employees, diagnose anyone, or make employment decisions. It helps people prepare for decisions; people make them.

The people in a decision

WethosAI represents people in three ways. Where the product uses an older name, it is shown in parentheses.

A Permissioned Twin (product label: Twin) is a permission-based AI model of a specific person. The person builds it themselves, chooses what it learns from, and decides whether it stays active. A Twin exists only with that person's permission. Twins: creating and training yours.
An Observed Twin (product label: Virtual Member) represents a specific person who has not created a Permissioned Twin, such as a board member, a client or a regulator. It is built from customer-approved organizational context and relevant publicly available professional information. It is less detailed than a Permissioned Twin, and it does not create one. Observed Twins (Virtual Members).
An Expert Agent (product label: Expert) is an AI agent with skills in a defined area of work, such as finance or security. It does not represent a real person. You add one when a decision needs a perspective nobody involved holds. Expert Agents (Utilizing Experts).

Every person who uses WethosAI also has a profile: a description of how they tend to take in information, weigh a decision and work with others. A profile is not a personality type and it is not a score. The four dimensions behind a profile.

Decision Simulation: rehearse before you act

A Decision Simulation (product label: Simulated Brainstorm) tests a decision against the people involved before you act. You describe the decision, attach the documents that matter, and choose who takes part: Permissioned Twins, Observed Twins and Expert Agents. The participants work through the question together. They raise questions, push back, look for tradeoffs and reach conclusions, and each conclusion comes with the reasoning behind it.

What you do inside a Decision Simulation is Decision Rehearsal: you try an approach, see where it meets resistance, change it, and run it again. Running a scenario more than once matters, because the patterns that recur across runs tell you more than any single run.

A Decision Simulation shows plausible responses and likely points of friction. It does not predict exactly what any real person will do. Treat what it shows as something to test with the real people, not as an answer.

Running a Decision Simulation.

Where the work happens

XO is the AI assistant inside WethosAI. It knows your profile, the documents and conversations in the space you are asking from, and the people you name. Ask it who needs to agree to a decision, what each of them is likely to question, and how to present it. Anything you say to XO on your own is private to you. Understanding and using XO.
A Brainstorm is a workspace for working through one decision with a chosen set of people, with XO taking part. A Decision Simulation runs inside a Brainstorm, and when it finishes the real people can join and carry on from what it found. A Breakout is a smaller conversation inside a Brainstorm. Brainstorms, Breakouts and Decision Simulations.
The Decisions Hub records each decision with the part of the conversation that produced it. The Decisions Hub.
A group is the workspace an admin sets up for a set of people, such as a leadership team or a project. Brainstorms live inside groups. Creating a Core Group.

An example

This is an illustration, not a customer story. A chief information security officer wants to know whether the approval step for urgent payment changes will hold up under pressure. They open a Brainstorm and run a Decision Simulation with two people in finance, an Observed Twin for the regional controller, and a security Expert Agent. They run the scenario several times, changing how urgent the request looks and whether the usual approver is available. In most runs, the approval holds. In the runs where the approver is unavailable, the request is escalated to someone without the context to challenge it. The team changes the escalation rule, runs the scenario again, and takes the revised control to the real people for review.

What WethosAI does not do

It does not predict exactly how a person will behave. It shows plausible responses so you can prepare.
It does not rate, rank or diagnose anyone, and it must not be used to make decisions about anyone's employment.
It does not create a Twin of anyone without their permission.
It does not act for anyone. Twins and Expert Agents support preparation; people make the decisions.
The people who manage your organization's account cannot read your private conversations with XO.

Read more: Privacy: who can see what. For any term on this page, Words we use has the definition.