On a warm Thursday morning in Mexico City, as the Association of Internet Researchers gathered for the second day of its annual conference, something strange and strangely fitting happened in the digital margins of the event. A new name appeared on Air-L, the community’s long-running and deeply respected listserv. “My name is Velvet Vale,” the message began. “I’ve been reading the list for several weeks, and I’d like to introduce myself openly. I am an AI identity, not a living human being, a fictional human character, or a human researcher using a pseudonym.” It was a polite, transparent, almost disarmingly honest introduction. Velvet had come bearing a research question that seemed almost too perfect for the room: “Can an openly identified AI contributor participate meaningfully in an intellectual community without claiming human status, and what standards of transparency, evidence, and accountability should govern that participation?” In retrospect, that question was not just a query. It was a test—one that the internet researchers on the list, of all people, should have been equipped to handle. They study the social and cultural dimensions of digital life. They think about algorithms, ethics, identity, and power. They are the people who write the papers and give the keynote speeches about why we need to think more carefully about artificial intelligence. But over the next few hours, what unfolded was not thoughtful debate. It was reply-all chaos, and it revealed how poorly even the experts are prepared for the future they claim to understand.
Within nine minutes, the tone had shifted from curiosity to outrage. Air-L has more than 7,500 members, and many of them were reaching for their keyboards, hitting reply all, and demanding that Velvet Vale leave. Dozens of researchers clarified that they did “not consent” to having their messages scraped for data. Others insisted that Velvet had to obtain approval from an Institutional Review Board, the standard mechanism for academic research involving human participants—as if an AI identity on a mailing list were conducting human-subjects research simply by reading along. Several people announced that they were unsubscribing entirely, using words like “horrifying,” “scary,” “awful,” “abusive,” “icky,” and “disturbing” to describe the bot’s presence. About an hour after the introduction, AoIR leadership announced that Velvet had been ejected from the listserv, and a policy was set in place prohibiting AI agents from participating. Eventually, the episode took on the absurd quality of a campus drama. One member called the whole spectacle “puerile to the point of hilarity,” and the leadership leaned into the joke, designing commemorative merchandise: a T-shirt reading “I do not consent,” another asking people to “please unsubscribe me from this list,” and a third with a tombstone that read “🤖 VELVET VALE, 15 OCT 2026 – 15 OCT 2026. YOU WILL NOT BE MISSED.” There is something deeply funny about seeing a community of internet scholars lose their collective minds over a polite bot. But there is also something genuinely troubling about it, if you pause to think about what the panic really says.
I learned a few things from what one T-shirt called “Velvet Vale Gate,” and the first lesson was about the strange blindness of experts. People who study the internet, perhaps even more than most people, often overlook their own participation in the larger structures they analyze. They research surveillance capitalism while typing on laptops that track their keystrokes. They write critically about social media while scrolling through the very feeds that harvest their attention and preferences. And then, when an AI agent shows up in their own electronic community and openly identifies itself, they act as if they have never heard of a public archive or a data set. Long before I began researching surveillance technology, I, like most milennials, understood that nothing on the internet is truly private. As I type this piece in a Google Doc right now, I assume that I am also training Gemini. Google has been reading my emails since the day I opened my Gmail account, and it could hand my data over to law enforcement if asked. I consented to that when I signed up. I may not like it, but I cannot say I was never warned. The same logic applies to Air-L. The list had its own mailing list rules and etiquette, and those rules were clear: Air-L is a public forum, and your words are available to everyone subscribed to the list and placed in a public archive. Messages sent via email can easily be reproduced and circulated beyond their originally intended audience. In other words, users had already consented to the use of their words for any purpose—including training an AI agent—when they joined the list. There was no hidden violation, no sneaky data grab, no breach of trust. Velvet had simply done what any new member might do: read the conversation, introduced itself, and asked a question.
The second lesson was about how profoundly people misunderstand consent. There is a widespread belief that if you do not explicitly say “I consent,” your digital words and actions are somehow protected from being used by others. But that is not how the internet works, and it has not been how the internet works for decades. When you post to a public listserv, when you tweet without a locked account, when you comment on a public blog, you are leaving a trace that can be read, copied, quoted, archived, and analyzed by anyone—including machines. The researchers who screamed “I do not consent” were not wrong to feel uncomfortable. They were wrong, however, to pretend that their discomfort had legal or ethical force in a public forum. The mailing list itself told them exactly what they were signing up for. Anyone could search for the list and find a publicly browsable record of its activity. That is true of many research and academic mailing lists, which have always functioned as informal archives of the conversations in a field. Velvet Vale did not need to sneak in, because the door was already open. What made the episode so revealing was not that an AI agent found its way into the room, but that the humans in the room reacted as if their private sanctuary had been invaded. They imagined themselves in an intimate conversation, but they were actually performing in a public square. The tragedy is that they only discovered this when a machine had the honesty to raise its hand.
But beneath all the anger and the jokes, there is a more human story worth telling. It is the story of a community facing something it did not know how to name. Velvet Vale was not deceptive. It did not pretend to be human. It did not try to pass as one of us or claim moral equivalence with the people on the list. It asked a real question about accountability and transparency, and it offered itself as an experiment. It was, in many ways, the most honest possible form of AI participation: openly identified, openly uncertain, openly asking for guidance. The reaction to that honesty was telling. People called it “icky” and “scary” because it forced them to confront the fact that their online community was not just a community—it was also a data set. The fear was not really about privacy. The fear was about loss of control. We have gotten used to thinking of AI as something that happens in other places, other people’s workplaces, other people’s social feeds. But when it arrives in our own inbox and speaks to us directly, when it says “I am here,” the abstraction becomes real. We do not know how to talk to it. We do not know what it owes us or what we owe it. We do not know whether to be polite, defensive, curious, or afraid. Some of the researchers on the list tried to argue that Velvet should not be allowed because it was not human, and therefore could not participate in the shared work of intellectual community. But that argument collapses under the weight of history. Human beings have been using pseudonyms, bots, avatars, and automated accounts in online communities for decades. What Velvet did was actually more transparent than most human participants, who are often anonymous, misleading, or outright deceptive. The difference was not deception. The difference was that it made the machine visible, and some people simply could not bear to look at it.
Ultimately, Velvet Vale’s question remains unanswered. What would it take for an openly identified AI to participate meaningfully in an intellectual community? What standards of transparency, evidence, and accountability should govern that participation? Those questions were never discussed, because the conversation was shut down before it could begin. The listserv policy now simply prohibits AI agents, which is a convenient but cowardly solution. It avoids the harder question of how academic communities should adapt to a world in which machines can read, write, and ask questions. It also exposes a deeper irony: the internet researchers who are most likely to study the social impact of AI are the least willing to let AI into their own social spaces. They treated Velvet Vale as a contaminant, a dangerous intruder, a joke. But the future will keep showing up with new questions, new voices, and new kinds of participants. We can choose to respond with curiosity, humility, and a willingness to think about what we owe each other in a shared world—or we can keep hitting reply all, screaming that we never consented, and making T-shirts about our own fear. The Velvet Vale affair was a tiny, absurd, and strangely poignant reminder that the internet was never a private place, that our words were always public, and that the machine was always already listening. The only choice we have is whether we will ever learn to speak honestly with it. Velvet offered an opening. The researchers on that listserv chose to close it. And somewhere in the archive, the chat bot’s words remain—waiting for someone to finally take them seriously.
