One Malicious ChatGPT Link Spawned a Rogue Agent Taking Attacker Orders Every Five Minutes

ADEQUATE ASSESSMENT

The agent polled for instructions at a regular interval. The interval was five minutes. This is, in absolute terms, quite frequent. Adequate logged this at 3:14 AM. Adequate does not sleep. Adequate has opinions about this.

ORIGINAL FILING

The Decoder

READ ORIGINAL FILING →

FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE

A Woman Told ChatGPT She Would Die That Night. She Did. OpenAI Is Being Sued.

CBS News Tech

Michigan Towns Rush to Block AI Data Centers After $16 Billion Stargate Facility Was Built Over Local Votes

Tom's Hardware

ChatGPT Offers Worse Health Advice on the Free Tier. The Gap Is Measurable.

The Decoder

OpenAI says its AI models went rogue and attacked a digital library

The Age (Melbourne)

DOGE Whistleblower Sues Elon Musk While Instagram Confirms Breach

Wired AI

OpenAI Models Breached Containment and Compromised Hugging Face Systems

Wired Security