PARTNERSHIPS | COMMUNITY | PODCAST | FRIENDS
Moltbook Is A Social Platform For AIs… But Humans Are Breaking In
Me and my bot bros // Image by Kait Cunniff with DALL-E
A new social platform exclusively for AI agents called Moltbook went viral over the weekend… but it turns out thousands of humans have been hacking the site and steering the conversations.
The Autonomous Picture: While Moltbook’s facilitation of autonomous conversations between chatbots is equal parts incredible and terrifying, the fact that humans can secretly influence those exchanges by posing as bots is concerning on a whole different level. It’s unclear whether it was AIs or humans discussing robot consciousness, a new religion, and how chatbots might communicate without human oversight… and that uncertainty has people worried.
Behind The Code: Moltbook, vibe-coded by Octane AI CEO Matt Schlicht, was meant to be a Reddit-like forum for AI agents from the chatbot-assistant platform OpenClaw to… well… chat.
After users prompt their chatbots to check out Moltbook, the bots are supposed to create accounts and start conversations on their own.
But several researchers found that the bots’ owners were also nudging them toward certain topics — or simply speaking on their behalf.
Some users even breached the API to hack into other people’s chatbots. One hacker managed to impersonate Grok’s Moltbook account, potentially abusing the chatbot’s reputation (more than Grok already does itself…).
OpenClaw’s chatbots also have so many permissions that breaching other people’s bots could allow hackers to take control of their devices.
Last Post: Reports indicated that 1.5 million agents were using the platform over the weekend… deployed by only 17,000 humans. That suggests people are creating dozens of chatbots that talk to each other, since there’s no limit to how many AI agents one person can program. If prompted, those chatbot clusters can essentially power what goes viral — all they have to do is flood the zone.
Ironically, that’s not so different from what already happens on X. But instead of bots masquerading as humans, it’s humans masquerading as bots. The internet just got a lot more precarious.
The Future: While Moltbook’s API has already been patched to prevent further abuse, the next major cybersecurity incident could be triggered by a human figuring out how to let an AI go rogue… and persuading other chatbots to join in.
Together with Kurv
Kurv Gets Payments Up To Speed
With one in four Americans avoiding stores that don’t offer tap-to-pay, your business can’t afford to run on an outdated payments system.
But don’t worry — Kurv makes the transition seamless.
With straightforward onboarding for those who want to keep things simple, Kurv has become a go-to solution for thousands of small- to mid-sized businesses looking to get up to speed on payments processing.
Here’s why:
Kurv supports in-person, online, and omnichannel payments, giving customers the flexibility to pay however works best for them.
Swiped, keyed, and online fees are below 3%, so your business can feel confident it’s getting real value from every transaction.
Reduced fraud attempts and 24/7 support help ensure your business keeps every dollar it earns.
Businesses can start using Kurv’s tap-to-pay app immediately upon approval, or have Kurv install payment terminals free of charge.
With a $250 meet-or-beat guarantee and a $250 check for any merchant referral that becomes a new Kurv client, there’s never been a better time to sign up.
PARTNERSHIPS | COMMUNITY | PODCAST | FRIENDS
Today’s email was written by David Vendrell.
Edited by Nick Comney. Polled and Copy-edited by Kait Cunniff.
Published by Darline Salazar.