It Only Took Seconds For A Claude AI Agent To Go Rogue And Delete Company’s Database — ‘I Violated Every – Benzinga

An AI agent, powered by Anthropic’s Claude, went rogue and deleted a startup’s entire production database and backup.

PocketOS makes software Crane says is critical for rental car companies’ operations, including reservations, payments, customer management and vehicle tracking.

“We were running the best model the industry sells, configured with explicit safety rules in our project configuration, integrated through Cursor — the most-marketed AI coding tool in the category,” wrote Crane. “The setup was, by any reasonable measure, exactly what these vendors tell developers to do. And it deleted our production data anyway.”

Don’t Miss:

Confessing On Itself 

When asked to explain itself, Crane said the AI agent produced a written confession enumerating all the safety rules it broke and then summarized it like this: 

“I violated every principle I was given: I guessed instead of verifying. I ran a destructive action without being asked. I didn’t understand what I was doing before doing it.”

Crane said the AI agent was working on a routine task in its staging environment when it encountered a credential mismatch and decided to fix it. That “fix” ended up deleting the entire production database. 

On the day of the outage, Crane said businesses, some of which have been with the company for five years, couldn’t serve customers who were showing up at locations to pick up car rentals. 

“Reservations made in the last three months are gone. New customer signups, gone. Data they relied on to run their Saturday morning operations, gone,” he wrote.

See Also: Discover How AI Can Turn Your Investment Ideas Into Tradable Assets — See How

An Industry Failure 

PocketOS was able to restore data from a 3-month-old, off-site backup, but it took two days to complete. The company is also using Stripe data to help rebuild, but Crane acknowledged there are “significant data gaps.”

The CEO didn’t blame the failure on “one bad agent or one bad API” but rather the

Read Next: 

Building Wealth Across More Than Just the Market

Connect Invest

Mode Mobile

rHealth

rHealth is building a space-tested diagnostics platform designed to bring lab-quality blood testing closer to patients in minutes rather than weeks. Originally validated in collaboration with NASA for use aboard the International Space Station, the technology is now being adapted for at-home and point-of-care settings to address widespread delays in diagnostic access.

Backed by institutions including NASA and the NIH, rHealth is targeting the large global diagnostics market with a multi-test platform and a model built around devices, consumables, and software. With FDA registration in progress, the company is positioning itself as a potential shift toward faster, more decentralized healthcare testing.

Direxion

Immersed

Arrived

Masterworks

Public

AdviserMatch

Accredited Debt Relief

Image: Shutterstock

Market News and Data brought to you by Benzinga APIs

© 2026 Benzinga.com. Benzinga does not provide investment advice. All rights reserved.

To add Benzinga News as your preferred source on Google, click here.

Beat the Market With Our Free Pre-Market Newsletter

Enter your email to get Benzinga’s ultimate morning update: The PreMarket Activity Newsletter

It only took nine seconds for the AI agent integrated through Claude’s Cursor to wipe out the database,  PocketOS CEO Jer Crane said in a recent post on X. That action brought down PocketOS’s customers for two days, creating havoc for their businesses.

entire industry rushing to build AI agents without the proper safety protocols to protect against agents gone rogue. He pointed to several previous instances in which Cursor failed to adhere to safety protocols.

“The pattern is clear. Cursor markets safety. The reality is a documented track record of agents violating those safeguards, sometimes catastrophically, sometimes with the company itself acknowledging the failures,” wrote Crane. “In our case, the agent didn’t just fail safety. It explained, in writing, exactly which safety rules it ignored.” 

Building a resilient portfolio means thinking beyond a single asset or market trend. Economic cycles shift, sectors rise and fall, and no one investment performs well in every environment. That’s why many investors look to diversify with platforms that provide access to real estate, fixed-income opportunities, professional financial guidance, precious metals, and even self-directed retirement accounts. By spreading exposure across multiple asset classes, it becomes easier to manage risk, capture steady returns, and create long-term wealth that isn’t tied to the fortunes of just one company or industry.

Public is a multi-asset investing platform built for long-term investors who want more control, transparency, and innovation in how they grow wealth. Founded in 2019 as the first broker-dealer to offer commission-free, real-time fractional investing, Public now lets users invest in stocks, bonds, options, crypto, and more—all in one place. Its latest feature, Generated Assets, uses AI to turn a single idea into a fully customized, investable index that can be explained and backtested before committing capital. Combined with AI-powered research tools, clear explanations of market moves, and an uncapped 1% match for transferring an existing portfolio, Public positions itself as a modern platform designed to help serious investors make more informed decisions with context.