OpenAI just published a new 37-page report on how bad actors are attempting to misuse ChatGPT

Some of the wild cases:

- A fraud ring scaled personalized romance scams with AI-generated scripts

- North Korea-linked actors used it to research crypto attack vectors and draft fake recruiter messages

- A single operator ran political influence campaigns across 6+ countries on Facebook, Instagram, and X at the same time

- State-backed hackers from China, Iran, and North Korea used it for recon and phishing

- One ring built a full "scam-as-a-service" operation, translating fraud scripts into Japanese, generating fake job listings, and crafting SMS lures

None of this is new. It's just 10x faster now, and likely only going to get more sophisticated and harder to detect.