Recently, the CEOs of a major video conferencing company and a well-known fintech company each let an AI avatar of themselves stand in for part of a public company update. Not the whole thing — both disclosed clearly that they were speaking with an avatar, and the avatar covered only a portion of the meeting before the actual person took over. Still, the reaction was swift, and pointed.
The concern wasn't really about that one moment. It was about what it signals. If a CEO's avatar can credibly handle part of a high-stakes update today, the worry goes, how much more of leadership — of any human role — does it handle next year? That's the real question underneath the headlines: not "was this specific use OK," but "where does this end."
It's worth saying plainly: this wasn't a unanimous verdict. Plenty of people were genuinely impressed — the delivery was polished, the disclosure was upfront, and for some, it read as a reasonable, transparent use of the technology for a defined, limited purpose. The sharper reaction came from a different place: a sense that this is an early data point in a much longer trend, and that trend is people being present less, even when they're still nominally "there."
Both things can be true at once. A single, disclosed, partial use of an avatar isn't itself a problem. And it's still fair to ask what happens as the practice spreads, gets used for more of the meeting, and eventually stops needing a disclosure at all because it's just assumed.
We don't think the answer is "never use an avatar for anything leadership says." We also don't think the answer is "it's fine because they disclosed it, case closed." The honest answer is that this is a moving target, and it deserves an ongoing conversation rather than a one-time ruling — from the companies building this technology, and from the leaders deciding how to use it.
Here's where we land, for now: the deciding factor isn't whether an avatar was used. It's whether a human being was actually directing it — deciding what got said, reviewing it before it went out, and standing behind it afterward. That's the line we build around: Humans Direct. AI Executes. Humans Sign Off.
An avatar covering part of a meeting, with disclosure, with the real person present for the rest of it, and with every word approved in advance — that's a genuinely different thing from an avatar quietly doing more and more of the talking until no one's quite sure how much of "the leader" is actually in the room anymore. The technology doesn't draw that line by itself. People have to.
We don't have a tidy answer for exactly how far this goes, and we're skeptical of anyone who claims they do. What we do think is that the companies experimenting with this — and the ones like us building the tools for it — owe people a clear, honest account of who's actually in charge each time it's used. That's a conversation we intend to keep having, not a box we're checking once.
We'd genuinely like to hear where you land on this. If you've got a reaction — impressed, uneasy, somewhere in between — we'd love to hear it.
