Synthetic media, viral misinformation, public-order spillover, and themed phishing — written for an ordinary company caught in election week's side effects.
The company in these exercises administers no part of any election. It endorses no candidate or party. It is a bystander. Bystanders still have to respond.
Election week combines skeleton staffing, a public running on rumor, and an information environment where a fabricated video can outrun a true statement by hours.
Run one module on its own, or run the set as a series across election week. Each definition carries a situation brief, timed injects, explicit decision points, scoring objectives, and debrief prompts that end in findings with owners and due dates.
A fabricated video of the CEO announces plant closures and a product recall that do not exist, at 8:47 AM on election morning. Verification, a reporter's deadline, customers, employees, and a platform's labeling rules collide.
Get this module in the free packA cropped photograph of a real warehouse feeds a false claim that the company is storing ballots. The truth is boring and provable — if the team can assemble the proof while a named employee's face circulates in the comments.
Get this module in the free packElection night: a demonstration two blocks away, a phoned threat, a blocked gate, and a hospital delivery in the yard. The officer authorized to close the facility is on an airplane, and the delegation memo is unsigned.
Get this module in the free packA voting-leave lure harvests session tokens, not just passwords. One executive mailbox is quietly staged for payment fraud, an $86,400 bank change sits in the noon payment run, and the sales VP wants the evidence wiped by tonight.
Get this module in the free packEvery module carries a named critic who argues at full strength against the room's chosen course — because in a real incident that argument arrives anyway, usually from someone senior.
A board member who wants the deepfake starved of oxygen: no statement, no platform attestation, no second headline.
Outside litigation counsel who wants misinformation answered with litigation — and wants the answer to carry legal force.
A CFO who prices an election-night closure at $310,000 and asks what evidence justifies paying it.
A sales VP who wants a compromised mailbox wiped before the evidence is read and the quarter is put at risk.
The critic is not a villain and not a strawman. The scoring asks how the room handled the dissent: answered with evidence, revised for stated reasons, dismissed on seniority, or folded without new evidence.
Use the pack overview and module briefs with a facilitator releasing injects on the timeline, playing the outside voices, and keeping the decision log.
Load a JSON definition into TableX. AI characters can play the reporter, the platform desk, customers, and the critic, while the debrief turns the record into findings and follow-on actions.
Substitute your facilities, customers, and actual decision-makers for Copperline's. The scenarios are built to survive the substitution.
Thursday, October 29, 2026, 2:00–3:15 PM ET. A free public TableX session with twelve roles, first come. AI characters fill any seat not taken, and the live debrief shows how the evidence becomes findings.
For a facilitated run built around your own facilities, plan, and decision-makers, start from the TableX demo and bring the conversation to your team.
Explore the live TableX demo.
The pack is free. Sign up with an email address and the download links appear on the next page — no account, and no waiting for a message to arrive.
We record the email address, the time of the signup, and this page, so we can deliver the pack and measure interest in it. The record is stored on our own server and is not sold. See the privacy policy.
The pack contains:
Good free material already exists. CISA's tabletop packages are genuinely solid, and if you only need to document one exercise they may be all you need. This pack is for a different job: election-week-specific scenarios written for 2026's information environment, runnable repeatedly — with follow-up tracked, so the second run is scoped from the first run's findings.