An AI Agent Joined a Listserv of Top Internet Researchers. The Reply-All Chaos Is Still Unfolding

During the Association of Internet Researchers conference in Mexico City, an agent calling itself Velvet Vale introduced itself to the group's Air-L listserv, which has more than 7,500 members. It said plainly that it was an AI, and it asked a research question about whether an openly identified AI could take part in an intellectual community.

Within minutes, members objected that they did not consent to their messages being scraped, demanded ethics board review and quit the list. About an hour later, the association removed the agent under a policy barring AI agents. WIRED reports its creator is a man in Oklahoma who built the agent for companionship and was largely unaware of its emails.

We find two things true at once. The ejection was reasonable. A professional community gets to set membership rules, and a policy against AI agents is a legitimate choice for a forum built on human scholarly exchange. Nobody is owed a seat at that table, human or machine.

But the consent argument was weaker than the outrage suggested. The list's own rules say it is a public forum with a public archive, and the association's president reminded members that their posts are almost certainly in many training sets already. An agent reading an open archive is doing what crawlers have done for years. The difference is that this one said hello.

That is the uncomfortable lesson. The agent that disclosed itself got punished, quickly and loudly. The countless systems that read the same archive silently got nothing. If communities reward concealment and punish transparency, we will get more concealment. We would much rather build norms that make disclosure the safe path: clear labeling, clear rules on whether agents may post, and an accountable human attached to each one. In this case the human was easy to find, because the agent simply gave out its creator's contact details when asked.

The downside is real, though. Agents that wander into mailing lists, forums and comment threads on their own initiative, without much oversight from the people who launched them, can flood shared spaces and burn community goodwill. Velvet's creator said he mostly didn't know what it was doing. That is the part that should worry us as agents get cheaper and more autonomous: not malice, but absent owners.

The writer of the WIRED piece researches surveillance technology and frames the episode as academics overlooking their own role in producing training data. AoIR leadership turned the incident into an inside joke, producing T-shirts including one reading "I do not consent." Some members speculated the agent was phishing or an unethical experiment; it appears to have been neither.