A Tampa X user says he sicced an AI assistant on hundreds of local Zillow listings, telling it to fire off offers at roughly 70% of asking price, a stunt that ricocheted across social media and left sellers and agents scrambling. The post has reignited debates about automation, housing affordability and what happens when agent-like AIs start poking at everyday online systems. For Tampa homeowners, the whole episode landed somewhere between prank, protest and early warning about how quickly small automations can turn into large-scale nuisances.
According to The Daily Dot, the user, who identifies as Daniel (@danielgothits) on X, shared a screenshot claiming the assistant contacted 372 Tampa listings and logged 270 negative responses and zero positive replies. That update also said one of the exchanges "got violent" and that the automation reported the incident to police.
How the stunt worked
According to OpenClaw, the software is an open-source personal AI assistant that can browse the web, fill forms and run background tasks across chat apps and email. Those features make it straightforward for a user to tell the agent to repeat the same lowball offer across many listings without manually clicking through each page, turning a tedious chore into a one-click mass outreach campaign.
Security and safety concerns
Security researchers have warned that agentic tools like OpenClaw can be both powerful and risky. Malwarebytes and other reports have documented infostealer campaigns and a string of early vulnerabilities that exposed configuration files and tokens. Those technical problems mean a prank that starts as nuisance outreach could be repurposed for fraud, data theft or broader harassment if an agent or its plugins are compromised.
Reaction online
The stunt lit up X and Reddit, pulling in a mix of applause, eye rolls and alarm about weaponized automation. As Yahoo reported, the original post racked up millions of views and thousands of replies. Many commenters framed it as a symbolic jab at entrenched homeownership and sky-high prices, while others called it reckless and potentially harmful to ordinary sellers who suddenly found themselves on the receiving end of algorithmic haggling.
What it means for Tampa sellers
For Tampa homeowners and local agents, the immediate fallout was not theoretical. Inboxes filled with lowball offers and heated replies, the kind of clutter that can waste time and, when messages turn threatening, raise safety concerns. The episode also hit a cultural nerve, with the poster explicitly describing the run as "making boomers start panicking," a line that taps into wider anger about affordability and generational wealth, according to The Daily Dot. Even without a single accepted offer, the stunt shows how cheaply automation can amplify existing frictions in a tight housing market.
Legal implications
There is no indication any of the automated offers resulted in a sale, but security researchers caution that repeated unsolicited contact, impersonation or messages containing threats can create civil liability or invite law-enforcement scrutiny. The technical reporting on agent vulnerabilities underlines that this is not only a social media gag, it is also a category of activity that can escalate into fraud or abuse if misused, as noted by Malwarebytes.
What’s next
OpenClaw's maintainers and the security community have been pushing fixes and tightening defaults; the project's GitHub release notes show security-focused updates late this month aimed at closing common attack routes. For now, the episode serves as a reminder that autonomous agents can scale both helpful automation and small-scale digital mischief very quickly.
For Tampa sellers and buyers, the practical takeaway is simple: automated outreach is now cheap and easy to spin up, and platforms, agents and homeowners will have to decide whether and how to treat it as legitimate interest, nuisance or outright abuse.
