Ben Guez, a startup founder and content creator, has automated his dating life using OpenClaw, Claude Code, and Instagram’s trial reels feature. He told TechCrunch he now has “a bunch of potential international wives in [his] DMs” as a result.

How the Automation Works

The pipeline is straightforward. OpenClaw monitors World Cup match results. After each game, the agent triggers Claude Code to generate an Instagram trial reel using a fixed template: Guez staring out a train window looking dejected, captioned “I can’t believe {COUNTRY} lost… If any {COUNTRY} girls need emotional support… my DMs are open.”

Trial reels don’t appear on a creator’s public profile, so Guez has posted the same template more than a dozen times without any of them visible to casual profile visitors. The results: over one million views and 200 DMs in a few days, according to TechCrunch.

There’s a secondary funnel built in. Guez’s profile states he only responds to DMs sent through Canary, his AI language learning app. Every respondent has to download it first.

”They’re More Impressed”

Guez told TechCrunch that the women who discover the automation aren’t angry: “They’re more impressed, like, ‘Oh, you’re thinking outside of the box, you’re a genius.’” TechCrunch noted it was unable to independently verify those reactions.

The Line Between Automation and Manipulation

Not everyone using OpenClaw for dating is running the same playbook. Jeff Weisbein, founder of a tech PR firm in South Florida, told TechCrunch he uses OpenClaw to research restaurants and plan date logistics across different neighborhoods. He draws the line at mediating actual conversations.

“I have seen people create bots and ways to swipe using OpenClaw, and I wouldn’t do that,” Weisbein said. “I feel like you shouldn’t delegate your communication when you’re in a relationship with someone to AI.”

The pattern reflects a broader split in how personal AI agents are being used. Logistics and research automation, where agents replace Google searches and spreadsheet work, carries different ethical weight than agents that manage social interactions on a user’s behalf without the other party’s knowledge.

What makes Guez’s setup notable for the agent community isn’t the dating angle. It’s the consent architecture, or lack of one. The women responding to his trial reels don’t know they’re interacting with an automated outreach system. The reel looks personal. The country reference looks timely. The emotional framing looks genuine. All of it is templated and machine-generated.

As AI agents move from business workflows into personal and social contexts, the question of disclosure becomes harder to dodge. Business automation has norms around it. Personal manipulation through agents operating without the other party’s awareness does not.