Logged
Every agent action appears in the record's Activity feed under Agent, distinct from User, System and Rule events. Each agent keeps a run history.
AI event management: agents that do the work, under your approval
AI event management in Sessionboard means a set of supervised AI agents working on your real event record. Reviewer pre-scores submissions against your review plan, Coordinator chases speakers with open tasks, Scheduler drafts the agenda, Editor polishes session and speaker copy, and Ask answers questions about the event in plain language. Every action is logged, scoped and reversible, so your team keeps every decision.
QUICK ANSWER
What are AI agents for event management?
AI agents for event management are supervised assistants that act on your program data: in Sessionboard, Reviewer, Coordinator, Scheduler, Editor and Ask each take one job that used to take a person days. They propose, you approve, and every step lands in the record's Activity feed.
REVIEWER
Pre-score every submission before the committee opens it
The Reviewer reads session submissions and scores them against your review plan's scorecard, with the same rating scales and comment fields your reviewers see. It works through personas, AI reviewers with a name, a background and a review style, default or your own, up to three active on an event. Scores and summaries land next to your human reviewers' scores on the plan round, your committee overrides anything it disagrees with, and the run history shows every submission it processed.
- Scorecard of your review plan
- Up to three personas
- Default or custom personas
- Scores beside human reviewers
- Run history
COORDINATOR
Chase speakers with open tasks, in your tone
Speakers owe bios, headshots, slides and consent forms, and someone has to ask three times. The Coordinator finds speakers with open portal tasks, drafts the follow-up in the cadence and tone you set, and shows you the drafts before anything is sent. Each run records who was contacted and what went out, under Agent in the speaker's Activity feed.
- Cadence and tone
- Open portal tasks
- Drafts you review
- Sent log per run
SCHEDULER
Draft the agenda and read why each session landed there
Give the Scheduler a draft agenda's window, session durations and the rules that matter (keynotes in the main hall, no speaker in two rooms at once) and it proposes a schedule. Each placement comes with the reason it was made, so the program team can argue with the draft instead of building it, and you commit only the changes you accept into the live agenda.
- Window, durations, rules
- Reasoned placements
- Conflict-free by construction
- Accept per change
EDITOR
Polish session and speaker copy to your brand voice
Titles, descriptions and bios arrive in every register. The Editor reviews the fields you choose against the tone and brand voice you set and suggests a rewrite for each one, accepted or dismissed from the record it belongs to, so the published program reads as one voice without anyone retyping it.
- Tone and brand voice
- Fields you choose
- Accept or dismiss each rewrite
ASK AND TEAM LEAD
Ask anything about the event, on any page
Ask opens on any event page and tells you where you are, what is already set up there and what to ask next. Full screen in the Agent window as Team Lead, you ask in plain words, read answers set like a document, open records, dashboards and reports beside the conversation, and approve every change before it lands. The same window exists for an organization.
- Page guide
- Agent window
- Records beside the chat
- Approve every change
- Organization-wide
HOW TRUST WORKS
Supervised autonomy, built into every agent
Agents build on the same data and permissions as the rest of Sessionboard. What an agent can see and do is governed by its scopes and your Event Team permissions.
Scoped
Agents inherit token scopes and your organization's PII settings. Nothing an agent does exceeds what the person who enabled it could do.
Reversible
Agent changes are auditable in History and can be rolled back. Reviewer scores, Scheduler placements and Editor rewrites all wait for a person.
Yours to run
Each agent is a switch on the Agents page. Off until you turn it on, and it only works on what you assign to it: a plan round, a draft agenda, a set of fields.
SPEAKER CRM
Agents in the pipeline, too
In the Speaker CRM pipeline, agents research and enrich prospects so the sourcing board fills itself with context: who this person is, where they have spoken and what they talk about. The work is logged in the card's Activity feed under Agent, the same way as everything else.
- Prospect research
- Enrichment
- Logged in the card
YOUR OWN ASSISTANT
Or bring Claude, ChatGPT, Copilot or Gemini
The agents above run inside Sessionboard. If your team already works in Claude, ChatGPT, Microsoft Copilot or Google Gemini, connect it through the Sessionboard MCP server with OAuth and ask about sessions, speakers and reports on the same record, with personal data masked in query results and writes gated by the scopes you grant. The Sessionboard app for Slack answers in the channel your team already uses.
- MCP server
- OAuth sign-in
- Masked query results
- Slack app
5
agents on the Agents page
3
Reviewer personas per event
1
Activity feed for every action
0
changes without approval
AI event management, answered
What is AI event management?
AI event management is using AI agents inside your event platform to do the repetitive work of a program: pre-scoring submissions, chasing speakers for missing items, drafting the agenda, polishing copy and answering questions about the event. In Sessionboard the agents act on your real event record and every action is logged, scoped and reversible, so your team keeps the decisions.
Which AI agents does Sessionboard ship?
Reviewer (pre-scores submissions against your review plan through personas), Coordinator (drafts follow-ups to speakers with open tasks), Scheduler (proposes a draft agenda and explains each placement), Editor (suggests rewrites to session and speaker copy), and Ask, which answers in plain language and opens full screen in the Agent window as Team Lead. Agents also enrich prospects in the Speaker CRM pipeline.
Do the agents make decisions without us?
No. Agents run at supervised autonomy: they propose and act within the scopes you give them, and the changes that matter wait for your approval. Reviewer scores sit next to your human reviewers' scores and the committee still decides. Scheduler commits only the changes you accept. Editor rewrites are accepted or dismissed one at a time.
How do we know what an agent did?
Every agent action appears in the record's Activity feed under Agent, distinct from User, System and Rule events, and each agent has a run history that shows what it processed, what it produced and how the run went. Changes are auditable in History and can be rolled back.
Can an agent see personal data?
Agents inherit the same permissions as the rest of Sessionboard: token scopes, Event Team permissions and your organization's PII settings. An agent cannot see or change anything the person who enabled it could not.
Does the Reviewer replace our review committee?
No. The Reviewer reads every submission first and scores it on the same scorecard your reviewers use, through up to three personas you choose or write. Your committee starts from a considered opinion instead of a blank form, overrides anything it disagrees with, and makes the acceptance decision.
Can we use our own AI assistant instead?
Yes. Claude, ChatGPT, Microsoft Copilot, Google Gemini or any MCP-compatible assistant connects through the Sessionboard MCP server with OAuth, and the Sessionboard app for Slack answers in the channels your team already uses, on the same record the agents work on.