What a 500-Person AI Webinar Can and Can't Do
On February 5, 2026, I delivered an AI awareness webinar for a global advertising group in Hong Kong. About five hundred people on the call -- firmwide, the kind of session a global company runs when it decides everyone should have a baseline understanding of what AI tools do.
Five hundred is a number worth putting on a slide. It's a less useful number for measuring whether anyone's work looked different by Wednesday.
What a broadcast does
A five-hundred-person webinar does one thing well: it puts a floor under awareness. Before the session, some portion of the company has never opened an AI tool. After, everyone has at least seen the tools used in real time, heard the reasonable applications, and had someone address the questions that surface in every corporate audience. That has real value. Shared vocabulary matters -- when marketing, finance, and operations can reference the same demonstration, the conversation about what to do next starts from a common base instead of from five hundred separate internet searches.
A firmwide session also signals permission. In most organisations, a meaningful number of employees aren't sure whether using AI tools at work is encouraged, tolerated, or quietly frowned upon. A company-wide webinar organized by leadership answers that without anyone having to ask. The signal -- the company considers this important enough to put everyone in a room for it -- is itself the message.
What a broadcast doesn't
Nobody changes their workflow because of a webinar. The format doesn't allow meaningful hands-on time -- not when you can't see screens or troubleshoot individual problems. Questions go into the chat and get answered generically. The session ends, and by Tuesday most people have returned to whatever tools they were already using. Broadcast creates awareness at a scope nothing else matches, and awareness without follow-up structure decays at the rate it always has.
What happens after a session like this: a fraction of the audience tries something on their own. A smaller fraction keeps at it past the first week. Everyone else needs a next step -- smaller sessions, hands-on workshops, structured follow-up -- to move from knowing what AI is to using it in their actual work. A good awareness webinar generates that demand. Expecting it to also fulfill it is the mistake I see most often.
Three design choices that make the broadcast worth running
First, demonstrate with the company's own material. The distance between "AI can summarize a document" and "AI just summarized the kind of report you write every week" is the distance between interesting and relevant. Generic examples are safe. The audience's actual work is what holds attention.
Second, the live component has to be live. Pre-recorded demos are polished and convincingly ignorable. A live demonstration where the tool might produce something unexpected -- and occasionally does -- keeps five hundred people paying attention because the outcome is genuinely uncertain.
Third, hold real time for questions. The questions that surface at the end of a firmwide session are usually better than anything in the prepared material, because they come from people already thinking about their own work. Those questions -- about language support, about reliability, about whether this replaces their role -- are worth more to the room than any planned slide.
Months after the global advertising group session, Adaptig was asked to quote a follow-up for the same client. We declined. The proposed scope didn't include enough follow-up structure to produce a result we'd stand behind. I would rather lose a booking than deliver a session I don't think will change anything. That costs revenue. It's also why the sessions I do take tend to generate repeat bookings -- which, over time, is where the work actually comes from.
